WO2025029609A1

WO2025029609A1 - Personalization techniques for media playback systems

Info

Publication number: WO2025029609A1
Application number: PCT/US2024/039698
Authority: WO
Inventors: Matthew BENATAN
Original assignee: Sonos Inc
Current assignee: Sonos Inc
Priority date: 2023-07-28
Filing date: 2024-07-26
Publication date: 2025-02-06
Anticipated expiration: 2026-01-28

Abstract

Techniques for personalizing one or more settings of playback devices in a media playback system. In one example, a method includes providing collected sample values of one or more settings of one or more playback devices, and feature data associated with the sample values, to a parameterized machine learning model to train the model to predict a recommended value of the one or more settings using the sample values and the feature data, based on current feature data associated with at least one of the playback devices, applying the parameterized machine learning model to the current feature data to generate the recommended value of the one or more settings, and causing at least one of the playback devices to perform at least one playback device operation based on the recommended value.

Description

PERSONALIZATION TECHNIQUES FOR MEDIA PLAYBACK SYSTEMS

CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to co-pending U.S. Provisional Application No. 63/516,343 titled “PERSONALIZATION TECHNIQUES FOR MEDIA PLAYBACK SYSTEM” and filed on July 28, 2023, which is incorporated herein by reference in its entirety for all purposes.

FIELD OF THE DISCLOSURE

[0002] The present disclosure is related to consumer goods and, more particularly, to methods, systems, products, aspects, services, and other elements directed to media playback or some aspect thereof.

BACKGROUND

[0003] Options for accessing and listening to digital audio in an out-loud setting were limited until in 2002, when Sonos, Inc. began development of a new type of playback system. Sonos then filed one of its first patent applications in 2003, entitled “Method for Synchronizing Audio Playback between Multiple Networked Devices,” and began offering its first media playback systems for sale in 2005. The SONOS Wireless Home Sound System enables people to experience music from many sources via one or more networked playback devices. Through a software control application installed on a controller (e.g., smartphone, tablet, computer, voice input device), one can play what she wants in any room having a networked playback device. Media content (e.g., songs, podcasts, video sound) can be streamed to playback devices such that each room with a playback device can play back corresponding different media content. In addition, rooms can be grouped together for synchronous playback of the same media content, and/or the same media content can be heard in all rooms synchronously.

BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Various aspects, and advantages of the presently disclosed technology may be better understood with regard to the following description, appended claims, and accompanying drawings, as listed below. A person skilled in the relevant art will understand that the attributes shown in the drawings are for purposes of illustrations, and variations, including different and/or additional attributes and arrangements thereof, are possible. [0005] Figure 1 A is a partial cutaway view of an environment having a media playback system configured in accordance with aspects of the disclosed technology.

[0006] Figure IB is a schematic diagram of the media playback system of Figure 1A and one or more networks.

[0007] Figure 1C is a block diagram of a playback device.

[0008] Figure ID is a block diagram of a playback device.

[0009] Figure IE is a block diagram of a bonded playback device.

[0010] Figure IF is a block diagram of a network microphone device.

[0011] Figure 1G is a block diagram of a playback device.

[0012] Figure 1H is a partial schematic diagram of a control device.

[0013] Figures II through IL are schematic diagrams of corresponding media playback system zones.

[0014] Figure IM is a schematic diagram of media playback system areas.

[0015] Figure 2A is a front isometric view of a playback device configured in accordance with aspects of the disclosed technology.

[0016] Figure 2B is a front isometric view of the playback device of Figure 2A without a grille.

[0017] Figure 2C is an exploded view of the playback device of Figure 2A.

[0018] Figure 3A is a front view of a network microphone device configured in accordance with aspects of the disclosed technology.

[0019] Figure 3B is a side isometric view of the network microphone device of Figure 3 A.

[0020] Figure 3C is an exploded view of the network microphone device of Figures 3 A and 3B.

[0021] Figure 3D is an enlarged view of a portion of Figure 3B.

[0022] Figure 3E is a block diagram of the network microphone device of Figures 3A-3D.

[0023] Figure 3F is a schematic diagram of an example voice input.

[0024] Figures 4A-4D are schematic diagrams of a control device in various stages of operation in accordance with aspects of the disclosed technology.

[0025] Figure 5 is a message flow diagram of a media playback system.

[0026] Figure 6 is a conceptual diagram illustrating aspects of a positioning system architecture in accordance with aspects of the disclosure.

[0027] Figure 7 is a block diagram of one example of a system including a personalization service in accordance with aspects of the present disclosure.

[0028] Figure 8 is a flow diagram of one example of a process implemented by a model selector in accordance with aspects of the present disclosure. [0029] Figure 9 is a flow diagram of one example of a process producing a personalization recommendation based on a sample of input data in accordance with aspects of the present disclosure.

[0030] Figure 10 is a graph illustrating one example of volume data for a household in accordance with aspects of the present disclosure.

[0031] Figure 11 is a sequence diagram of one example of a process for producing a volume personalization recommendation in accordance with aspects of the present disclosure.

[0032] Figure 12 is a diagram illustrating an example of movement of a portable playback device within an environment in accordance with aspects of the present disclosure.

[0033] Figure 13 is a diagram illustrating another example of movement of a portable playback device within an environment in accordance with aspects of the present disclosure.

[0034] Figure 14 is a diagram illustrating an example of a user-requested personalization configuration in accordance with aspects of the present disclosure.

[0035] Figure 15 A is a sequence diagram of one example of a process for producing a grouping personalization recommendation in accordance with aspects of the present disclosure.

[0036] Figure 15B is a sequence diagram of one example of a personalization process incorporating user confirmation in accordance with aspects of the present disclosure.

[0037] Figure 16 is a block diagram of one example of a system including a personalization service in accordance with aspects of the present disclosure.

[0038] The drawings are for the purpose of illustrating example embodiments, but those of ordinary skill in the art will understand that the technology disclosed herein is not limited to the arrangements and/or instrumentality shown in the drawings.

DETAILED DESCRIPTION

I. Overview

[0039] Embodiments described herein relate to techniques for personalizing a user experience with a media playback system. Many users demonstrate consistent listening routines or patterns when using capabilities and/or devices within their media playback system. According to certain aspects, techniques are provided for predicting user preferences based on detecting one or more consistent patterns over time. For example, historical usage data and/or recorded patterns of movement can be used to train machine learning models that can then automatically adjust, or prompt a user to adjust, certain settings or configurations of one of more playback devices in the media playback system. For example, in the context of volume personalization, the system may determine that a user routinely sets the volume of a certain playback device to 20% and selects a certain radio station on weekday mornings. Based on determining this pattern, the system can predict the user’s preferences for the playback device configuration. Thus, through personalization, the system can reduce the time and user effort required to achieve the predicted end result (e.g., the “time to music”) by automatically setting the volume to 20% and selecting the certain radio station when the user activates that playback device on weekday mornings.

[0040] In some examples, the system can be configured to automatically apply personalization settings based on recognized patterns or routines. In other examples, the system can be configured to offer suggestions for playback device configuration settings to the user based on the recognized pattern(s), or request confirmation from the user to correlate a particular routine with a playback device configuration. According to certain examples, the machine learning models can incorporate confidence indicators (e.g., representing the level of confidence the system has regarding the correlation between a detected repeated pattern of movement and/or behavior and certain device or system configurations that are affected by the pattern, and therefore the likelihood that the user may want this routine automated) that can be used as a factor in determining the action to be taken by the system in a given scenario. For example, if the model has a high level of confidence associated with the correlation between a particular routine and a playback device configuration, the system may automatically apply the personalization settings. Alternatively, if the confidence level is lower, the system may instead request user confirmation and/or offer suggested personalization settings to the user, rather than automatically implementing the predicted personalization settings. In certain examples, the models can incorporate learning based on user feedback (positive or negative) regarding personalization settings or suggestions, and adjust system personalization behavior towards improving the user experience with the media playback system.

[0041] In some embodiments, for example, a method of personalizing a setting of one or more playback devices in a media playback system includes collecting, over time, sample values of the setting and feature data associated with the sample values. As used herein, the term “feature data” refers to data specifying or describing one or more features extracted from collected data that is input to a parameterized machine learning model. This collected input data may include the sample values of the setting, for example, and other data (e.g., context data, sample values of other settings of one or more playback devices, etc.), as described further below. In some examples, at least one element of the feature data is derived from one or more of a time at which a corresponding sample value was collected or a location of the one or more playback devices when the correspond sample value was collected. Examples of the method further comprise training a parameterized machine learning model to predict a recommended value of the setting using the sample values and the feature data. The setting may include, for example, a volume setting of one or more of the playback devices or grouping behavior for the one or more playback devices. Examples of the method further include detecting context data representative of one or more of a current time, a current user interaction with the one or more playback devices, or current location of the one or more playback devices extracting current feature data from the context data, and applying the parameterized machine learning model to the current feature data to generate the recommended value of the setting and a confidence metric corresponding to the recommended value. A playback device operation, such as automatically adjusting the setting or recommending an adjustment of the setting to a user, may then be executed based on the recommended value and the confidence metric.

[0042] While some examples described herein may refer to functions performed by given actors such as “users,” “listeners,” and/or other entities, it should be understood that this is for purposes of explanation only. The claims should not be interpreted to require action by any such example actor unless explicitly required by the language of the claims themselves.

[0043] In the Figures, identical reference numbers identify generally similar, and/or identical, elements. To facilitate the discussion of any particular element, the most significant digit or digits of a reference number refers to the Figure in which that element is first introduced. For example, element 110a is first introduced and discussed with reference to Figure 1 A. Many of the details, dimensions, angles, and other characteristics shown in the Figures are merely illustrative of particular embodiments of the disclosed technology. Accordingly, other embodiments can have other details, dimensions, angles, and characteristics without departing from the spirit or scope of the disclosure. In addition, those of ordinary skill in the art will appreciate that further embodiments of the various disclosed technologies can be practiced without several of the details described below.

II. Suitable Operating Environment

[0044] Figure 1A is a partial cutaway view of a media playback system 100 distributed in an environment 101 (e.g., a house). The media playback system 100 comprises one or more playback devices 110 (identified individually as playback devices HOa-n), one or more network microphone devices 120 (“NMDs”) (identified individually as NMDs 120a-c), and one or more control devices 130 (identified individually as control devices 130a and 130b).

[0045] As used herein the term “playback device” can generally refer to a network device configured to receive, process, and output data of a media playback system. For example, a playback device can be a network device that receives and processes audio content. In some embodiments, a playback device includes one or more transducers or speakers powered by one or more amplifiers. In other embodiments, however, a playback device includes one of (or neither of) the speaker and the amplifier. For instance, a playback device can comprise one or more amplifiers configured to drive one or more speakers external to the playback device via a corresponding wire or cable.

[0046] Moreover, as used herein the term “NMD” (i.e., a “network microphone device”) can generally refer to a network device that is configured for audio detection. In some embodiments, an NMD is a stand-alone device configured primarily for audio detection. In other embodiments, an NMD is incorporated into a playback device (or vice versa).

[0047] The term “control device” can generally refer to a network device configured to perform functions relevant to facilitating user access, control, and/or configuration of the media playback system 100.

[0048] Each of the playback devices 110 is configured to receive audio signals or data from one or more media sources (e.g., one or more remote servers, one or more local devices, etc.) and play back the received audio signals or data as sound. The one or more NMDs 120 are configured to receive spoken word commands, and the one or more control devices 130 are configured to receive user input. In response to the received spoken word commands and/or user input, the media playback system 100 can play back audio via one or more of the playback devices 110. In certain embodiments, the playback devices 110 are configured to commence playback of media content in response to a trigger. For instance, one or more of the playback devices 110 can be configured to play back a morning playlist upon detection of an associated trigger condition (e.g., presence of a user in a kitchen, detection of a coffee machine operation, etc.). In some embodiments, for example, the media playback system 100 is configured to play back audio from a first playback device (e.g., the playback device 100a) in synchrony with a second playback device (e.g., the playback device 100b). Interactions between the playback devices 110, NMDs 120, and/or control devices 130 of the media playback system 100 configured in accordance with the various embodiments of the disclosure are described in greater detail below with respect to Figures IB-5.

[0049] In the illustrated embodiment of Figure 1 A, the environment 101 comprises a household having several rooms, spaces, and/or playback zones, including (clockwise from upper left) a master bathroom 101a, a master bedroom 101b, a second bedroom 101c, a family room or den 101 d, an office lOle, a living room lOlf, a dining room 101g, a kitchen lOlh, and an outdoor patio lOli. While certain embodiments and examples are described below in the context of a home environment, the technologies described herein may be implemented in other types of environments. In some embodiments, for example, the media playback system 100 can be implemented in one or more commercial settings (e.g., a restaurant, mall, airport, hotel, a retail or other store), one or more vehicles (e.g., a sports utility vehicle, bus, car, a ship, a boat, an airplane, etc.), multiple environments (e.g., a combination of home and vehicle environments), and/or another suitable environment where multi-zone audio may be desirable.

[0050] The media playback system 100 can comprise one or more playback zones, some of which may correspond to the rooms in the environment 101. The media playback system 100 can be established with one or more playback zones, after which additional zones may be added, or removed, to form, for example, the configuration shown in Figure 1A. Each zone may be given a name according to a different room or space such as the office lOle, master bathroom 101a, master bedroom 101b, the second bedroom 101c, kitchen lOlh, dining room 101g, living room 10 If, and/or the balcony lOli. In some aspects, a single playback zone may include multiple rooms or spaces. In certain aspects, a single room or space may include multiple playback zones.

[0051] In the illustrated embodiment of Figure 1A, the master bathroom 101a, the second bedroom 101c, the office lOle, the living room 10 If, the dining room 101g, the kitchen lOlh, and the outdoor patio lOli each include one playback device 110, and the master bedroom 101b and the den 101 d include a plurality of playback devices 110. In the master bedroom 101b, the playback devices 1101 and 110m may be configured, for example, to play back audio content in synchrony as individual ones of playback devices 110, as a bonded playback zone, as a consolidated playback device, and/or any combination thereof. Similarly, in the den 101 d, the playback devices HOh-j can be configured, for instance, to play back audio content in synchrony as individual ones of playback devices 110, as one or more bonded playback devices, and/or as one or more consolidated playback devices. Additional details regarding bonded and consolidated playback devices are described below with respect to Figures IB, IE and II - IM.

[0052] In some aspects, one or more of the playback zones in the environment 101 may each be playing different audio content. For instance, a user may be grilling on the patio lOli and listening to hip hop music being played by the playback device 110c while another user is preparing food in the kitchen lOlh and listening to classical music played by the playback device 110b. In another example, a playback zone may play the same audio content in synchrony with another playback zone. For instance, the user may be in the office lOle listening to the playback device 1 lOf playing back the same hip hop music being played back by playback device 110c on the patio lOli. In some aspects, the playback devices 110c and 11 Of play back the hip hop music in synchrony such that the user perceives that the audio content is being played seamlessly (or at least substantially seamlessly) while moving between different playback zones. Additional details regarding audio playback synchronization among playback devices and/or zones can be found, for example, in U.S. Patent No. 8,234,395 entitled, “System and method for synchronizing operations among a plurality of independently clocked digital data processing devices,” which is incorporated herein by reference in its entirety. a. Suitable Media Playback System

[0053] Figure IB is a schematic diagram of the media playback system 100 and a cloud network 102. For ease of illustration, certain devices of the media playback system 100 and the cloud network 102 are omitted from Figure IB. One or more communication links 103 (referred to hereinafter as “the links 103”) communicatively couple the media playback system 100 and the cloud network 102.

[0054] The links 103 can comprise, for example, one or more wired networks, one or more wireless networks, one or more wide area networks (WAN), one or more local area networks (LAN), one or more personal area networks (PAN), one or more telecommunication networks (e.g., one or more Global System for Mobiles (GSM) networks, Code Division Multiple Access (CDMA) networks, Long-Term Evolution (LTE) networks, 5G communication networks, and/or other suitable data transmission protocol networks), etc. The cloud network 102 is configured to deliver media content (e.g., audio content, video content, photographs, social media content, etc.) to the media playback system 100 in response to a request transmitted from the media playback system 100 via the links 103. In some embodiments, the cloud network 102 is further configured to receive data (e.g., voice input data) from the media playback system 100 and correspondingly transmit commands and/or media content to the media playback system 100.

[0055] The cloud network 102 comprises computing devices 106 (identified separately as a first computing device 106a, a second computing device 106b, and a third computing device 106c). The computing devices 106 can comprise individual computers or servers, such as, for example, a media streaming service server storing audio and/or other media content, a voice service server, a social media server, a media playback system control server, etc. In some embodiments, one or more of the computing devices 106 comprise modules of a single computer or server. In certain embodiments, one or more of the computing devices 106 comprise one or more modules, computers, and/or servers. Moreover, while the cloud network 102 is described above in the context of a single cloud network, in some embodiments the cloud network 102 comprises a plurality of cloud networks comprising communicatively coupled computing devices. Furthermore, while the cloud network 102 is shown in Figure IB as having three of the computing devices 106, in some embodiments, the cloud network 102 comprises fewer (or more than) three computing devices 106.

[0056] The media playback system 100 is configured to receive media content from the networks 102 via the links 103. The received media content can comprise, for example, a Uniform Resource Identifier (URI) and/or a Uniform Resource Locator (URL). For instance, in some examples, the media playback system 100 can stream, download, or otherwise obtain data from a URI or a URL corresponding to the received media content. A network 104 communicatively couples the links 103 and at least a portion of the devices (e.g., one or more of the playback devices 110, NMDs 120, and/or control devices 130) of the media playback system 100. The network 104 can include, for example, a wireless network (e.g., a WI-FI network, a BLUETOOTH, a Z-WAVE network, a ZIGBEE, and/or other suitable wireless communication protocol network) and/or a wired network (e.g., a network comprising Ethernet, Universal Serial Bus (USB), and/or another suitable wired communication). As those of ordinary skill in the art will appreciate, as used herein, “WI-FI” can refer to several different communication protocols including, for example, Institute of Electrical and Electronics Engineers (IEEE) 802.11a, 802.11b, 802.11g, 802.1 In, 802.1 lac, 802. Had, 802.11af, 802.11 ah, 802.1 lai, 802.1 laj, 802.11aq, 802.1 lax, 802.1 lay, 802.15, etc. transmitted at 2.4 Gigahertz (GHz), 5 GHz, and/or another suitable frequency.

[0057] In some embodiments, the network 104 comprises a dedicated communication network that the media playback system 100 uses to transmit messages between individual devices and/or to transmit media content to and from media content sources (e.g., one or more of the computing devices 106). In certain embodiments, the network 104 is configured to be accessible only to devices in the media playback system 100, thereby reducing interference and competition with other household devices. In other embodiments, however, the network 104 comprises an existing household or commercial facility communication network (e.g., a household or commercial facility WI-FI network). In some embodiments, the links 103 and the network 104 comprise one or more of the same networks. In some aspects, for example, the links 103 and the network 104 comprise a telecommunication network (e.g., an LTE network, a 5G network, etc.). Moreover, in some embodiments, the media playback system 100 is implemented without the network 104, and devices comprising the media playback system 100 can communicate with each other, for example, via one or more direct connections, PANs, telecommunication networks, and/or other suitable communication links. The network 104 may be referred to herein as a “local communication network” to differentiate the network 104 from the cloud network 102 that couples the media playback system 100 to remote devices, such as cloud servers that host cloud services.

[0058] In some embodiments, audio content sources may be regularly added or removed from the media playback system 100. In some embodiments, for example, the media playback system 100 performs an indexing of media items when one or more media content sources are updated, added to, and/or removed from the media playback system 100. The media playback system 100 can scan identifiable media items in some or all folders and/or directories accessible to the playback devices 110, and generate or update a media content database comprising metadata (e.g., title, artist, album, track length, etc.) and other associated information (e.g., URIs, URLs, etc.) for each identifiable media item found. In some embodiments, for example, the media content database is stored on one or more of the playback devices 110, network microphone devices 120, and/or control devices 130.

[0059] In the illustrated embodiment of Figure IB, the playback devices 1101 and 110m comprise a group 107a. The playback devices 1101 and 110m can be positioned in different rooms and be grouped together in the group 107a on a temporary or permanent basis based on user input received at the control device 130a and/or another control device 130 in the media playback system 100. When arranged in the group 107a, the playback devices 1101 and 110m can be configured to play back the same or similar audio content in synchrony from one or more audio content sources. In certain embodiments, for example, the group 107a comprises a bonded zone in which the playback devices 1101 and 110m comprise left audio and right audio channels, respectively, of multi-channel audio content, thereby producing or enhancing a stereo effect of the audio content. In some embodiments, the group 107a includes additional playback devices 110. In other embodiments, however, the media playback system 100 omits the group 107a and/or other grouped arrangements of the playback devices 110. Additional details regarding groups and other arrangements of playback devices are described in further detail below with respect to Figures II through IM.

[0060] The media playback system 100 includes the NMDs 120a and 120b, each comprising one or more microphones configured to receive voice utterances from a user. In the illustrated embodiment of Figure IB, the NMD 120a is a standalone device and the NMD 120b is integrated into the playback device 1 lOn. The NMD 120a, for example, is configured to receive voice input 121 from a user 123. In some embodiments, the NMD 120a transmits data associated with the received voice input 121 to a voice assistant service (VAS) configured to (i) process the received voice input data and (ii) facilitate one or more operations on behalf of the media playback system 100.

[0061] In some aspects, for example, the computing device 106c comprises one or more modules and/or servers of a VAS (e.g., a VAS operated by one or more of SONOS, AMAZON, GOOGLE, APPLE, MICROSOFT, etc.). The computing device 106c can receive the voice input data from the NMD 120a via the network 104 and the links 103.

[0062] In response to receiving the voice input data, the computing device 106c processes the voice input data (i.e., “Play Hey Jude by The Beatles”), and determines that the processed voice input includes a command to play a song (e.g., “Hey Jude”). In some embodiments, after processing the voice input, the computing device 106c accordingly transmits commands to the media playback system 100 to play back “Hey Jude” by the Beatles from a suitable media service (e.g., via one or more of the computing devices 106) on one or more of the playback devices 110. In other embodiments, the computing device 106c may be configured to interface with media services on behalf of the media playback system 100. In such embodiments, after processing the voice input, instead of the computing device 106c transmitting commands to the media playback system 100 causing the media playback system 100 to retrieve the requested media from a suitable media service, the computing device 106c itself causes a suitable media service to provide the requested media to the media playback system 100 in accordance with the user’s voice utterance. b. Suitable Playback Devices

[0063] Figure 1C is a block diagram of the playback device 110a comprising an input/output 111. The input/output 111 can include an analog I/O I l la (e.g., one or more wires, cables, and/or other suitable communication links configured to carry analog signals) and/or a digital I/O 11 lb (e.g., one or more wires, cables, or other suitable communication links configured to carry digital signals). In some embodiments, the analog I/O I l la is an audio line-in input connection comprising, for example, an auto-detecting 3.5mm audio line-in connection. In some embodiments, the digital EO 111b comprises a Sony/Philips Digital Interface Format (S/PDIF) communication interface and/or cable and/or a Toshiba Link (TOSLINK) cable. In some embodiments, the digital I/O 111b comprises a High-Definition Multimedia Interface (HDMI) interface and/or cable. In some embodiments, the digital EO 111b includes one or more wireless communication links comprising, for example, a radio frequency (RF), infrared, WI-FI, BLUETOOTH, or another suitable communication link. In certain embodiments, the analog EO I l la and the digital I/O 111b comprise interfaces (e.g., ports, plugs, jacks, etc.) configured to receive connectors of cables transmitting analog and digital signals, respectively, without necessarily including cables.

[0064] The playback device 110a, for example, can receive media content (e.g., audio content comprising music and/or other sounds) from a local audio source 105 via the input/output 111 (e.g., a cable, a wire, a PAN, a BLUETOOTH connection, an ad hoc wired or wireless communication network, and/or another suitable communication link). The local audio source 105 can comprise, for example, a mobile device (e.g., a smartphone, a tablet, a laptop computer, etc.) or another suitable audio component (e.g., a television, a desktop computer, an amplifier, a phonograph (such as an LP turntable), a Blu-ray player, a memory storing digital media files, etc.). In some aspects, the local audio source 105 includes local music libraries on a smartphone, a computer, a networked-attached storage (NAS), and/or another suitable device configured to store media files. In certain embodiments, one or more of the playback devices 110, NMDs 120, and/or control devices 130 comprise the local audio source 105. In other embodiments, however, the media playback system omits the local audio source 105 altogether. In some embodiments, the playback device 110a does not include an input/output 111 and receives all audio content via the network 104.

[0065] The playback device 110a further comprises electronics 112, a user interface 113 (e.g., one or more buttons, knobs, dials, touch-sensitive surfaces, displays, touchscreens, etc.), and one or more transducers 114 (referred to hereinafter as “the transducers 114”). The electronics 112 are configured to receive audio from an audio source (e.g., the local audio source 105) via the input/output 111 or one or more of the computing devices 106a-c via the network 104 (Figure IB), amplify the received audio, and output the amplified audio for playback via one or more of the transducers 114. In some embodiments, the playback device 110a optionally includes one or more microphones 115 (e.g., a single microphone, a plurality of microphones, a microphone array) (hereinafter referred to as “the microphones 115”). In certain embodiments, for example, the playback device 110a having one or more of the optional microphones 115 can operate as an NMD configured to receive voice input from a user and correspondingly perform one or more operations based on the received voice input.

[0066] In the illustrated embodiment of Figure 1C, the electronics 112 comprise one or more processors 112a (referred to hereinafter as “the processors 112a”), memory 112b, software components 112c, a network interface 112d, one or more audio processing components 112g (referred to hereinafter as “the audio components H2g”), one or more audio amplifiers 112h (referred to hereinafter as “the amplifiers 112h”), and power 112i (e.g., one or more power supplies, power cables, power receptacles, batteries, induction coils, Power-over Ethernet (POE) interfaces, and/or other suitable sources of electric power). In some embodiments, the electronics 112 optionally include one or more other components 112j (e.g., one or more sensors, video displays, touchscreens, battery charging bases, etc.).

[0067] The processors 112a can comprise clock-driven computing component(s) configured to process data, and the memory 112b can comprise a computer-readable medium (e.g., a tangible, non-transitory computer-readable medium loaded with one or more of the software components 112c) configured to store instructions for performing various operations and/or functions. The processors 112a are configured to execute the instructions stored on the memory 112b to perform one or more of the operations. The operations can include, for example, causing the playback device 110a to retrieve audio data from an audio source (e.g., one or more of the computing devices 106a-c (Figure IB)), and/or another one of the playback devices 110. In some embodiments, the operations further include causing the playback device 110a to send audio data to another one of the playback devices 110a and/or another device (e.g., one of the NMDs 120). Certain embodiments include operations causing the playback device 110a to pair with another of the one or more playback devices 110 to enable a multi-channel audio environment (e.g., a stereo pair, a bonded zone, etc.).

[0068] The processors 112a can be further configured to perform operations causing the playback device 110a to synchronize playback of audio content with another of the one or more playback devices 110. As those of ordinary skill in the art will appreciate, during synchronous playback of audio content on a plurality of playback devices, a listener will preferably be unable to perceive time-delay differences between playback of the audio content by the playback device 110a and the other one or more other playback devices 110. Additional details regarding audio playback synchronization among playback devices can be found, for example, in U.S. Patent No. 8,234,395, which was incorporated by reference above.

[0069] In some embodiments, the memory 112b is further configured to store data associated with the playback device 110a, such as one or more zones and/or zone groups of which the playback device 110a is a member, audio sources accessible to the playback device 110a, and/or a playback queue that the playback device 110a (and/or another of the one or more playback devices) can be associated with. The stored data can comprise one or more state variables that are periodically updated and used to describe a state of the playback device 110a. The memory 112b can also include data associated with a state of one or more of the other devices (e.g., the playback devices 110, NMDs 120, control devices 130) of the media playback system 100. In some aspects, for example, the state data is shared during predetermined intervals of time (e.g., every 5 seconds, every 10 seconds, every 60 seconds, etc.) among at least a portion of the devices of the media playback system 100, so that one or more of the devices have the most recent data associated with the media playback system 100.

[0070] The network interface 112d is configured to facilitate a transmission of data between the playback device 110a and one or more other devices on a data network such as, for example, the links 103 and/or the network 104 (Figure IB). The network interface 112d is configured to transmit and receive data corresponding to media content (e.g., audio content, video content, text, photographs) and other signals (e.g., non-transitory signals) comprising digital packet data including an Internet Protocol (IP)-based source address and/or an IP -based destination address. The network interface 112d can parse the digital packet data such that the electronics 112 properly receive and process the data destined for the playback device 110a.

[0071] In the illustrated embodiment of Figure 1C, the network interface 112d comprises one or more wireless interfaces 112e (referred to hereinafter as “the wireless interface 112e”). The wireless interface 112e (e.g., a suitable interface comprising one or more antennae) can be configured to wirelessly communicate with one or more other devices (e.g., one or more of the other playback devices 110, NMDs 120, and/or control devices 130) that are communicatively coupled to the network 104 (Figure IB) in accordance with a suitable wireless communication protocol (e.g., WI-FI, BLUETOOTH, LTE, etc.). In some embodiments, the network interface 112d optionally includes a wired interface 112f (e.g., an interface or receptacle configured to receive a network cable such as an Ethernet, a USB-A, USB-C, and/or Thunderbolt cable) configured to communicate over a wired connection with other devices in accordance with a suitable wired communication protocol. In certain embodiments, the network interface 112d includes the wired interface 112f and excludes the wireless interface 112e. In some embodiments, the electronics 112 exclude the network interface 112d altogether and transmits and receives media content and/or other data via another communication path (e.g., the input/output 111).

[0072] The audio components 112g are configured to process and/or filter data comprising media content received by the electronics 112 (e.g., via the input/output 111 and/or the network interface 112d) to produce output audio signals. In some embodiments, the audio processing components 112g comprise, for example, one or more digital-to-analog converters (DACs), audio preprocessing components, audio enhancement components, digital signal processors (DSPs), and/or other suitable audio processing components, modules, circuits, etc. In certain embodiments, one or more of the audio processing components 112g can comprise one or more subcomponents of the processors 112a. In some embodiments, the electronics 112 omit the audio processing components 112g. In some aspects, for example, the processors 112a execute instructions stored on the memory 112b to perform audio processing operations to produce the output audio signals.

[0073] The amplifiers 112h are configured to receive and amplify the audio output signals produced by the audio processing components 112g and/or the processors 112a. The amplifiers 112h can comprise electronic devices and/or components configured to amplify audio signals to levels sufficient for driving one or more of the transducers 114. In some embodiments, for example, the amplifiers 112h include one or more switching or class-D power amplifiers. In other embodiments, however, the amplifiers 112h include one or more other types of power amplifiers (e.g., linear gain power amplifiers, class-A amplifiers, class-B amplifiers, class- AB amplifiers, class-C amplifiers, class-D amplifiers, class-E amplifiers, class-F amplifiers, class- G amplifiers, class H amplifiers, and/or another suitable type of power amplifier). In certain embodiments, the amplifiers 112h comprise a suitable combination of two or more of the foregoing types of power amplifiers. Moreover, in some embodiments, individual ones of the amplifiers 112h correspond to individual ones of the transducers 114. In other embodiments, however, the electronics 112 include a single one of the amplifiers 112h configured to output amplified audio signals to a plurality of the transducers 114. In some other embodiments, the electronics 112 omit the amplifiers 112h.

[0074] The transducers 114 (e.g., one or more speakers and/or speaker drivers) receive the amplified audio signals from the amplifier 112h and render or output the amplified audio signals as sound (e.g., audible sound waves having a frequency between about 20 Hertz (Hz) and 20 kilohertz (kHz)). In some embodiments, the transducers 114 can comprise a single transducer. In other embodiments, however, the transducers 114 comprise a plurality of audio transducers. In some embodiments, the transducers 114 comprise more than one type of transducer. For example, the transducers 114 can include one or more low frequency transducers (e.g., subwoofers, woofers), mid-range frequency transducers (e.g., mid-range transducers, mid-woofers), and one or more high frequency transducers (e.g., one or more tweeters). As used herein, “low frequency” can generally refer to audible frequencies below about 500 Hz, “mid-range frequency” can generally refer to audible frequencies between about 500 Hz and about 2 kHz, and “high frequency” can generally refer to audible frequencies above 2 kHz. In certain embodiments, however, one or more of the transducers 114 comprise transducers that do not adhere to the foregoing frequency ranges. For example, one of the transducers 114 may comprise a mid-woofer transducer configured to output sound at frequencies between about 200 Hz and about 5 kHz. [0075] By way of illustration, Sonos, Inc. presently offers (or has offered) for sale certain playback devices including, for example, a “SONOS ONE,” “PLAY:1,” “PLAY:3,” “PLAY: 5,” “PLAYBAR,” “PLAYBASE,” “CONNECT: AMP,” “CONNECT,” “AMP,” “PORT,” and “SUB.” Other suitable playback devices may additionally or alternatively be used to implement the playback devices of example embodiments disclosed herein. Additionally, one of ordinary skill in the art will appreciate that a playback device is not limited to the examples described herein or to Sonos product offerings. In some embodiments, for example, one or more playback devices 110 comprise wired or wireless headphones (e.g., over-the-ear headphones, on-ear headphones, in-ear earphones, etc.). In other embodiments, one or more of the playback devices 110 comprise a docking station and/or an interface configured to interact with a docking station for personal mobile media playback devices. In certain embodiments, a playback device may be integral to another device or component such as a television, an LP turntable, a lighting fixture, or some other device for indoor or outdoor use. In some embodiments, a playback device omits a user interface and/or one or more transducers. For example, Figure. ID is a block diagram of a playback device 1 lOp comprising the input/output 111 and electronics 112 without the user interface 113 or transducers 114.

[0076] Figure IE is a block diagram of a bonded playback device HOq comprising the playback device 110a (Figure 1C) sonically bonded with the playback device HOi (e.g., a subwoofer) (Figure 1 A). In the illustrated embodiment, the playback devices 110a and 1 lOi are separate ones of the playback devices 110 housed in separate enclosures. In some embodiments, however, the bonded playback device HOq comprises a single enclosure housing both the playback devices 110a and HOi. The bonded playback device HOq can be configured to process and reproduce sound differently than an unbonded playback device (e.g., the playback device 110a of Figure 1C) and/or paired or bonded playback devices (e.g., the playback devices 1101 and 110m of Figure IB). In some embodiments, for example, the playback device 110a is a full-range playback device configured to render low frequency, midrange frequency, and high frequency audio content, and the playback device HOi is a subwoofer configured to render low frequency audio content. In some aspects, the playback device 110a, when bonded with the first playback device, is configured to render only the midrange and high frequency components of a particular audio content, while the playback device HOi renders the low frequency component of the particular audio content. In some embodiments, the bonded playback device HOq includes additional playback devices and/or another bonded playback device. Additional playback device embodiments are described in further detail below with respect to Figures 2A-3D. c. Suitable Network Microphone Devices (NMDs)

[0077] Figure IF is a block diagram of the NMD 120a (Figures 1 A and IB). The NMD 120a includes one or more voice processing components 124 (hereinafter “the voice components 124”) and several components described with respect to the playback device 110a (Figure 1C) including the processors 112a, the memory 112b, and the microphones 115. The NMD 120a optionally comprises other components also included in the playback device 110a (Figure 1C), such as the user interface 113 and/or the transducers 114. In some embodiments, the NMD 120a is configured as a media playback device (e.g., one or more of the playback devices 110), and further includes, for example, one or more of the audio components 112g (Figure 1C), the amplifiers 112h, and/or other playback device components. In certain embodiments, the NMD 120a comprises an Internet of Things (loT) device such as, for example, a thermostat, alarm panel, fire and/or smoke detector, etc. In some embodiments, the NMD 120a comprises the microphones 115, the voice processing components 124, and only a portion of the components of the electronics 112 described above with respect to Figure 1C. In some aspects, for example, the NMD 120a includes the processor 112a and the memory 112b (Figure 1C), while omitting one or more other components of the electronics 112. In some embodiments, the NMD 120a includes additional components (e.g., one or more sensors, cameras, thermometers, barometers, hygrometers, etc.).

[0078] In some embodiments, an NMD can be integrated into a playback device. Figure 1G is a block diagram of a playback device HOr comprising an NMD 120d. The playback device 11 Or can comprise many or all of the components of the playback device 110a and further include the microphones 115 and voice processing components 124 (Figure IF). The playback device 1 lOr optionally includes an integrated control device 130c. The control device 130c can comprise, for example, a user interface (e.g., the user interface 113 of Figure 1C) configured to receive user input (e.g., touch input, voice input, etc.) without a separate control device. In other embodiments, however, the playback device 11 Or receives commands from another control device (e.g., the control device 130a of Figure IB). Additional NMD embodiments are described in further detail below with respect to Figures 3 A-3F.

[0079] Referring again to Figure IF, the microphones 115 are configured to acquire, capture, and/or receive sound from an environment (e.g., the environment 101 of Figure 1A) and/or a room in which the NMD 120a is positioned. The received sound can include, for example, vocal utterances, audio played back by the NMD 120a and/or another playback device, background voices, ambient sounds, etc. The microphones 115 convert the received sound into electrical signals to produce microphone data. The voice processing components 124 receive and analyze the microphone data to determine whether a voice input is present in the microphone data. The voice input can comprise, for example, an activation word followed by an utterance including a user request. As those of ordinary skill in the art will appreciate, an activation word is a word or other audio cue signifying a user voice input. For instance, in querying the AMAZON VAS, a user might speak the activation word "Alexa." Other examples include "Ok, Google" for invoking the GOOGLE VAS and "Hey, Siri" for invoking the APPLE VAS.

[0080] After detecting the activation word, voice processing components 124 monitor the microphone data for an accompanying user request in the voice input. The user request may include, for example, a command to control a third-party device, such as a thermostat (e.g., NEST thermostat), an illumination device (e.g., a PHILIPS HUE lighting device), or a media playback device (e.g., a SONOS playback device). For example, a user might speak the activation word “Alexa” followed by the utterance “set the thermostat to 68 degrees” to set a temperature in a home (e.g., the environment 101 of Figure 1 A). The user might speak the same activation word followed by the utterance “turn on the living room” to turn on illumination devices in a living room area of the home. The user may similarly speak an activation word followed by a request to play a particular song, an album, or a playlist of music on a playback device in the home. Additional description regarding receiving and processing voice input data can be found in further detail below with respect to Figures 3A-3F. d. Suitable Control Devices

[0081] Figure 1H is a partial schematic diagram of the control device 130a (Figures 1A and IB). As used herein, the term “control device” can be used interchangeably with “controller” or “control system.” Among other attributes, the control device 130a is configured to receive user input related to the media playback system 100 and, in response, cause one or more devices in the media playback system 100 to perform an action(s) or operation(s) corresponding to the user input. In the illustrated embodiment, the control device 130a comprises a smartphone (e.g., an iPhone™ an Android phone, etc.) on which media playback system controller application software is installed. In some embodiments, the control device 130a comprises, for example, a tablet (e.g., an iPad™), a computer (e.g., a laptop computer, a desktop computer, etc.), and/or another suitable device (e.g., a television, an automobile audio head unit, an loT device, etc.). In certain embodiments, the control device 130a comprises a dedicated controller for the media playback system 100. In other embodiments, as described above with respect to Figure 1G, the control device 130a is integrated into another device in the media playback system 100 (e.g., one more of the playback devices 110, NMDs 120, and/or other suitable devices configured to communicate over a network).

[0082] The control device 130a includes electronics 132, a user interface 133, one or more speakers 134, and one or more microphones 135. The electronics 132 comprise one or more processors 132a (referred to hereinafter as “the processors 132a”), a memory 132b, software components 132c, and a network interface 132d. The processor 132a can be configured to perform functions relevant to facilitating user access, control, and configuration of the media playback system 100. The memory 132b can comprise data storage that can be loaded with one or more of the software components executable by the processor 132a to perform those functions. The software components 132c can comprise applications and/or other executable software configured to facilitate control of the media playback system 100. The memory 132b can be configured to store, for example, the software components 132c, media playback system controller application software, and/or other data associated with the media playback system 100 and the user.

[0083] The network interface 132d is configured to facilitate network communications between the control device 130a and one or more other devices in the media playback system 100, and/or one or more remote devices. In some embodiments, the network interface 132d is configured to operate according to one or more suitable communication industry standards (e.g., infrared, radio, wired standards including IEEE 802.3, wireless standards including IEEE 802.11a, 802.11b, 802.11g, 802.11n, 802.11ac, 802.15, 4G, LTE, etc.). The network interface 132d can be configured, for example, to transmit data to and/or receive data from the playback devices 110, the NMDs 120, other ones of the control devices 130, one of the computing devices 106 of Figure IB, devices comprising one or more other media playback systems, etc. The transmitted and/or received data can include, for example, playback device control commands, state variables, playback zone and/or zone group configurations. For instance, based on user input received at the user interface 133, the network interface 132d can transmit a playback device control command (e.g., volume control, audio playback control, audio content selection, etc.) from the control device 130a to one or more of the playback devices 110. The network interface 132d can also transmit and/or receive configuration changes such as, for example, adding/removing one or more playback devices 110 to/from a zone, adding/removing one or more zones to/from a zone group, forming a bonded or consolidated player, separating one or more playback devices from a bonded or consolidated player, among others. Additional description of zones and groups can be found below with respect to Figures II through IM. [0084] The user interface 133 is configured to receive user input and can facilitate control of the media playback system 100. The user interface 133 includes media content art 133a (e.g., album art, lyrics, videos, etc.), a playback status indicator 133b (e.g., an elapsed and/or remaining time indicator), media content information region 133c, a playback control region 133d, and a zone indicator 133e. The media content information region 133c can include a display of relevant information (e.g., title, artist, album, genre, release year, etc.) about media content currently playing and/or media content in a queue or playlist. The playback control region 133d can include selectable (e.g., via touch input and/or via a cursor or another suitable selector) icons to cause one or more playback devices in a selected playback zone or zone group to perform playback actions such as, for example, play or pause, fast forward, rewind, skip to next, skip to previous, enter/exit shuffle mode, enter/exit repeat mode, enter/exit cross fade mode, etc. The playback control region 133d may also include selectable icons to modify equalization settings, playback volume, and/or other suitable playback actions. In the illustrated embodiment, the user interface 133 comprises a display presented on a touch screen interface of a smartphone (e.g., an iPhone™ an Android phone, etc.). In some embodiments, however, user interfaces of varying formats, styles, and interactive sequences may alternatively be implemented on one or more network devices to provide comparable control access to a media playback system.

[0085] The one or more speakers 134 (e.g., one or more transducers) can be configured to output sound to the user of the control device 130a. In some embodiments, the one or more speakers comprise individual transducers configured to correspondingly output low frequencies, mid-range frequencies, and/or high frequencies. In some aspects, for example, the control device 130a is configured as a playback device (e.g., one of the playback devices 110). Similarly, in some embodiments the control device 130a is configured as an NMD (e.g., one of the NMDs 120), receiving voice commands and other sounds via the one or more microphones 135.

[0086] The one or more microphones 135 can comprise, for example, one or more condenser microphones, electret condenser microphones, dynamic microphones, and/or other suitable types of microphones or transducers. In some embodiments, two or more of the microphones 135 are arranged to capture location information of an audio source (e.g., voice, audible sound, etc.) and/or configured to facilitate filtering of background noise. Moreover, in certain embodiments, the control device 130a is configured to operate as a playback device and an NMD. In other embodiments, however, the control device 130a omits the one or more speakers 134 and/or the one or more microphones 135. For instance, the control device 130a may comprise a device (e.g., a thermostat, an loT device, a network device, etc.) comprising a portion of the electronics 132 and the user interface 133 (e.g., a touch screen) without any speakers or microphones. Additional control device embodiments are described in further detail below with respect to Figures 4A-4D and 5. e. Suitable Playback Device Configurations

[0087] Figures II through IM show example configurations of playback devices in zones and zone groups. Referring first to Figure IM, in one example, a single playback device may belong to a zone. For example, the playback device 110g in the second bedroom 101c (Figure 1A) may belong to Zone C. In some implementations described below, multiple playback devices may be “bonded” to form a “bonded pair” which together form a single zone. For example, the playback device 1101 (e.g., a left playback device) can be bonded to the playback device 110m (e.g., a right playback device) to form Zone B. Bonded playback devices may have different playback responsibilities (e.g., channel responsibilities). In another implementation described below, multiple playback devices may be merged to form a single zone. For example, the playback device 1 lOh (e.g., a front playback device) may be merged with the playback device 1 lOi (e.g., a subwoofer), and the playback devices 1 lOj and 110k (e.g., left and right surround speakers, respectively) to form a single Zone D. In another example, the playback devices 110b and 1 lOd can be merged to form a merged group or a zone group 108b. The merged playback devices 110b and HOd may not be specifically assigned different playback responsibilities. That is, the merged playback devices 1 lOh and 1 lOi may, aside from playing audio content in synchrony, each play audio content as they would if they were not merged.

[0088] Each zone in the media playback system 100 may be provided for control as a single user interface (UI) entity. For example, Zone A may be provided as a single entity named Master Bathroom. Zone B may be provided as a single entity named Master Bedroom. Zone C may be provided as a single entity named Second Bedroom.

[0089] Playback devices that are bonded may have different playback responsibilities, such as responsibilities for certain audio channels. For example, as shown in Figure II, the playback devices 1101 and 110m may be bonded so as to produce or enhance a stereo effect of audio content. In this example, the playback device 1101 may be configured to play a left channel audio component, while the playback device 110m may be configured to play a right channel audio component. In some implementations, such stereo bonding may be referred to as “pairing.” [0090] Additionally, bonded playback devices may have additional and/or different respective speaker drivers. As shown in Figure 1 J, the playback device 1 lOh named Front may be bonded with the playback device 1 lOi named SUB. The Front device 1 lOh can be configured to render a range of mid to high frequencies and the SUB device 1 lOi can be configured to render low frequencies. When unbonded, however, the Front device I lOh can be configured to render a full range of frequencies. As another example, Figure IK shows the Front and SUB devices 11 Oh and HOi further bonded with Left and Right playback devices HOj and 110k, respectively. In some implementations, the Right and Left devices HOj and 102k can be configured to form surround or “satellite” channels of a home theater system. The bonded playback devices 1 lOh, 1 lOi, 1 lOj, and 110k may form a single Zone D (Figure IM).

[0091] Playback devices that are merged may not have assigned playback responsibilities, and may each render the full range of audio content the respective playback device is capable of. Nevertheless, merged devices may be represented as a single UI entity (i.e., a zone, as discussed above). For instance, the playback devices 110a and HOn in the master bathroom have the single UI entity of Zone A. In one embodiment, the playback devices 110a and 1 lOn may each output the full range of audio content each respective playback devices 110a and 11 On are capable of, in synchrony.

[0092] In some embodiments, an NMD is bonded or merged with another device so as to form a zone. For example, the NMD 120b may be bonded with the playback device I lOe, which together form Zone F, named Living Room. In other embodiments, a stand-alone network microphone device may be in a zone by itself. In other embodiments, however, a stand-alone network microphone device may not be associated with a zone. Additional details regarding associating network microphone devices and playback devices as designated or default devices may be found, for example, in subsequently referenced U.S. Patent No. 10,499,146.

[0093] Zones of individual, bonded, and/or merged devices may be grouped to form a zone group. For example, referring to Figure IM, Zone A may be grouped with Zone B to form a zone group 108a that includes the two zones. Similarly, Zone G may be grouped with Zone H to form the zone group 108b. As another example, Zone A may be grouped with one or more other Zones C-I. The Zones A-I may be grouped and ungrouped in numerous ways. For example, three, four, five, or more (e.g., all) of the Zones A-I may be grouped. When grouped, the zones of individual and/or bonded playback devices may play back audio in synchrony with one another, as described in previously referenced U.S. Patent No. 8,234,395. Playback devices may be dynamically grouped and ungrouped to form new or different groups that synchronously play back audio content. [0094] In various implementations, the zones in an environment may be the default name of a zone within the group or a combination of the names of the zones within a zone group. For example, Zone Group 108b can be assigned a name such as “Dining + Kitchen”, as shown in Figure IM. In some embodiments, a zone group may be given a unique name selected by a user.

[0095] Certain data may be stored in a memory of a playback device (e.g., the memory 112b of Figure 1C) as one or more state variables that are periodically updated and used to describe the state of a playback zone, the playback device(s), and/or a zone group associated therewith. The memory may also include the data associated with the state of the other devices of the media system, and shared from time to time among the devices so that one or more of the devices have the most recent data associated with the system.

[0096] In some embodiments, the memory may store instances of various variable types associated with the states. Variable instances may be stored with identifiers (e.g., tags) corresponding to type. For example, certain identifiers may be a first type “al” to identify playback device(s) of a zone, a second type “bl” to identify playback device(s) that may be bonded in the zone, and a third type “cl” to identify a zone group to which the zone may belong. As a related example, identifiers associated with the second bedroom 101c may indicate that the playback device is the only playback device of the Zone C and not in a zone group. Identifiers associated with the Den may indicate that the Den is not grouped with other zones but includes bonded playback devices 11 Oh- 110k. Identifiers associated with the Dining Room may indicate that the Dining Room is part of the Dining + Kitchen zone group 108b and that devices 110b and HOd are grouped (Figure IL). Identifiers associated with the Kitchen may indicate the same or similar information by virtue of the Kitchen being part of the Dining + Kitchen zone group 108b. Other example zone variables and identifiers are described below. [0097] In yet another example, the memory may store variables or identifiers representing other associations of zones and zone groups, such as identifiers associated with Areas, as shown in Figure IM. An area may involve a cluster of zone groups and/or zones not within a zone group. For instance, Figure IM shows an Upper Area 109a including Zones A-D and I, and a Lower Area 109b including Zones E-I. In one aspect, an Area may be used to invoke a cluster of zone groups and/or zones that share one or more zones and/or zone groups of another cluster. In another aspect, this differs from a zone group, which does not share a zone with another zone group. Further examples of techniques for implementing Areas may be found, for example, in U.S. Patent No. 10,712,997 filed August 21, 2017, and titled “Room Association Based on Name,” and U.S. Patent No. 8,483,853 filed September 11, 2007, and titled “Controlling and manipulating groupings in a multi-zone media system.” Each of these patents is incorporated herein by reference in its entirety. In some embodiments, the media playback system 100 may not implement Areas, in which case the system may not store variables associated with Areas.

111. Example Systems and Devices

[0098] Figure 2A is a front isometric view of a playback device 210 configured in accordance with aspects of the disclosed technology. Figure 2B is a front isometric view of the playback device 210 without a grille 216e. Figure 2C is an exploded view of the playback device 210. Referring to Figures 2A-2C together, the playback device 210 comprises a housing 216 that includes an upper portion 216a, a right or first side portion 216b, a lower portion, a left or second side portion 216d, the grille 216e, and a rear portion 216f. A plurality of fasteners 216g (e.g., one or more screws, rivets, clips) attaches a frame 216h to the housing 216. A cavity 216j (Figure 2C) in the housing 216 is configured to receive the frame 216h and electronics 212. The frame 216h is configured to carry a plurality of transducers 214 (identified individually in Figure 2B as transducers 214a-f). The electronics 212 (e.g., the electronics 112 of Figure 1C) is configured to receive audio content from an audio source and send electrical signals corresponding to the audio content to the transducers 214 for playback.

[0099] The transducers 214 are configured to receive the electrical signals from the electronics

112, and further configured to convert the received electrical signals into audible sound during playback. For instance, the transducers 214a-c (e.g., tweeters) can be configured to output high frequency sound (e.g., sound waves having a frequency greater than about 2 kHz). The transducers 214d-f (e.g., mid-woofers, woofers, midrange speakers) can be configured output sound at frequencies lower than the transducers 214a-c (e.g., sound waves having a frequency lower than about 2 kHz). In some embodiments, the playback device 210 includes a number of transducers different than those illustrated in Figures 2A-2C. For example, as described in further detail below with respect to Figures 3A-3C, the playback device 210 can include fewer than six transducers (e.g., one, two, three). In other embodiments, however, the playback device 210 includes more than six transducers (e.g., nine, ten). Moreover, in some embodiments, all or a portion of the transducers 214 are configured to operate as a phased array to desirably adjust (e.g., narrow or widen) a radiation pattern of the transducers 214, thereby altering a user’s perception of the sound emitted from the playback device 210.

[0100] In some examples, a filter is axially aligned with the transducer 214b. The filter can be configured to desirably attenuate a predetermined range of frequencies that the transducer 214b outputs to improve sound quality and a perceived sound stage output collectively by the transducers 214. In some embodiments, however, the playback device 210 omits the filter. In other embodiments, the playback device 210 includes one or more additional filters aligned with the transducers 214b and/or at least another of the transducers 214.

[0101] Figures 3A and 3B are front and right isometric side views, respectively, of an NMD 320 configured in accordance with embodiments of the disclosed technology. Figure 3C is an exploded view of the NMD 320. Figure 3D is an enlarged view of a portion of Figure 3B including a user interface 313 of the NMD 320. Referring first to Figures 3A-3C, the NMD 320 includes a housing 316 comprising an upper portion 316a, a lower portion 316b and an intermediate portion 316c (e.g., a grille). A plurality of ports, holes or apertures 316d in the upper portion 316a allow sound to pass through to one or more microphones 315 (Figure 3C) positioned within the housing 316. The one or more microphones 315 are configured to receive sound via the apertures 316d and produce electrical signals based on the received sound. In the illustrated embodiment, a frame 316e (Figure 3C) of the housing 316 surrounds cavities 316f and 316g configured to house, respectively, a first transducer 314a (e.g., a tweeter) and a second transducer 314b (e.g., a mid-woofer, a midrange speaker, a woofer). In other embodiments, however, the NMD 320 includes a single transducer, or more than two (e.g., two, five, six) transducers. In certain embodiments, the NMD 320 omits the transducers 314a and 314b altogether.

[0102] Electronics 312 (Figure 3C) includes components configured to drive the transducers 314a and 314b, and further configured to analyze audio data corresponding to the electrical signals produced by the one or more microphones 315. In some embodiments, for example, the electronics 312 comprises many or all of the components of the electronics 112 described above with respect to Figure 1C. In certain embodiments, the electronics 312 includes components described above with respect to Figure IF such as, for example, the one or more processors 112a, the memory 112b, the software components 112c, the network interface 112d, etc. In some embodiments, the electronics 312 includes additional suitable components (e.g., proximity or other sensors).

[0103] Referring to Figure 3D, the user interface 313 includes a plurality of control surfaces (e.g., buttons, knobs, capacitive surfaces) including a first control surface 313a (e.g., a previous control), a second control surface 313b (e.g., a next control), and a third control surface 313c (e.g., a play and/or pause control) that can be adjusted by a user 323. A fourth control surface 313d is configured to receive touch input corresponding to activation and deactivation of the one or microphones 315. A first indicator 313e (e.g., one or more light emitting diodes (LEDs) or another suitable illuminator) can be configured to illuminate only when the one or more microphones 315 are activated. A second indicator 313f (e.g., one or more LEDs) can be configured to remain solid during normal operation and to blink or otherwise change from solid to indicate a detection of voice activity. In some embodiments, the user interface 313 includes additional or fewer control surfaces and illuminators. In one embodiment, for example, the user interface 313 includes the first indicator 313e, omitting the second indicator 313f Moreover, in certain embodiments, the NMD 320 comprises a playback device and a control device, and the user interface 313 comprises the user interface of the control device.

[0104] Referring to Figures 3A-3D together, the NMD 320 is configured to receive voice commands from one or more adj acent users via the one or more microphones 315. As described above with respect to Figure IB, the one or more microphones 315 can acquire, capture, or record sound in a vicinity (e.g., a region within 10m or less of the NMD 320) and transmit electrical signals corresponding to the recorded sound to the electronics 312. The electronics 312 can process the electrical signals and can analyze the resulting audio data to determine a presence of one or more voice commands (e.g., one or more activation words). In some embodiments, for example, after detection of one or more suitable voice commands, the NMD 320 is configured to transmit a portion of the recorded audio data to another device and/or a remote server (e.g., one or more of the computing devices 106 of Figure IB) for further analysis. The remote server can analyze the audio data, determine an appropriate action based on the voice command, and transmit a message to the NMD 320 to perform the appropriate action. For instance, a user may speak “Sonos, play Michael Jackson.” The NMD 320 can, via the one or more microphones 315, record the user’s voice utterance, determine the presence of a voice command, and transmit the audio data having the voice command to a remote server (e.g., one or more of the remote computing devices 106 of Figure IB, one or more servers of a VAS and/or another suitable service). The remote server can analyze the audio data and determine an action corresponding to the command. The remote server can then transmit a command to the NMD 320 to perform the determined action (e.g., play back audio content related to Michael Jackson). The NMD 320 can receive the command and play back the audio content related to Michael Jackson from a media content source. As described above with respect to Figure IB, suitable content sources can include a device or storage communicatively coupled to the NMD 320 via a LAN (e.g., the network 104 of Figure IB), a remote server (e.g., one or more of the remote computing devices 106 of Figure IB), etc. In certain embodiments, however, the NMD 320 determines and/or performs one or more actions corresponding to the one or more voice commands without intervention or involvement of an external device, computer, or server.

[0105] Figure 3E is a functional block diagram showing additional attributes of the NMD 320 in accordance with aspects of the disclosure. The NMD 320 includes components configured to facilitate voice command capture including voice activity detector component(s) 312k, beam former components 3121, acoustic echo cancellation (AEC) and/or self-sound suppression components 312m, activation word detector components 312n, and voice/speech conversion components 312o (e.g., voice-to-text and text-to-voice). In the illustrated embodiment of Figure 3E, the foregoing components 312k-312o are shown as separate components. In some embodiments, however, one or more of the components 312k-312o are subcomponents of the processors 112a.

[0106] The beamforming and self-sound suppression components 3121 and 312m are configured to detect an audio signal and determine aspects of voice input represented in the detected audio signal, such as the direction, amplitude, frequency spectrum, etc. The voice activity detector activity components 312k are operably coupled with the beamforming and AEC components 3121 and 312m and are configured to determine a direction and/or directions from which voice activity is likely to have occurred in the detected audio signal. Potential speech directions can be identified by monitoring metrics which distinguish speech from other sounds. Such metrics can include, for example, energy within the speech band relative to background noise and entropy within the speech band, which is measure of spectral structure. As those of ordinary skill in the art will appreciate, speech typically has a lower entropy than most common background noise.

[0107] The activation word detector components 312n are configured to monitor and analyze received audio to determine if any activation words (e.g., wake words) are present in the received audio. The activation word detector components 312n may analyze the received audio using an activation word detection algorithm. If the activation word detector 312n detects an activation word, the NMD 320 may process voice input contained in the received audio. Example activation word detection algorithms accept audio as input and provide an indication of whether an activation word is present in the audio. Many first- and third-party activation word detection algorithms are known and commercially available. For instance, operators of a voice service may make their algorithm available for use in third-party devices. Alternatively, an algorithm may be trained to detect certain activation words. In some embodiments, the activation word detector 312n runs multiple activation word detection algorithms on the received audio simultaneously (or substantially simultaneously). As noted above, different voice services (e g., AMAZON’S ALEXA, APPLE’S SIRI, or MICROSOFT’S CORTANA) can each use a different activation word for invoking their respective voice service. To support multiple services, the activation word detector 312n may run the received audio through the activation word detection algorithm for each supported voice service in parallel.

[0108] The speech/text conversion components 312o may facilitate processing by converting speech in the voice input to text. In some embodiments, the electronics 312 can include voice recognition software that is trained to a particular user or a particular set of users associated with a household. Such voice recognition software may implement voice-processing algorithms that are tuned to specific voice profile(s). Tuning to specific voice profiles may require less computationally intensive algorithms than traditional voice activity services, which typically sample from a broad base of users and diverse requests that are not targeted to media playback systems.

[0109] Figure 3F is a schematic diagram of an example voice input 328 captured by the NMD 320 in accordance with aspects of the disclosure. The voice input 328 can include an activation word portion 328a and a voice utterance portion 328b. In some embodiments, the activation word 328a can be a known activation word, such as “Alexa,” which is associated with AMAZON'S ALEXA. In other embodiments, however, the voice input 328 may not include an activation word. In some embodiments, a network microphone device may output an audible and/or visible response upon detection of the activation word portion 328a. In addition, or alternately, an NMD may output an audible and/or visible response after processing a voice input and/or a series of voice inputs.

[0110] The voice utterance portion 328b may include, for example, one or more spoken commands (identified individually as a first command 328c and a second command 328e) and one or more spoken keywords (identified individually as a first keyword 328d and a second keyword 328f) . In one example, the first command 328c can be a command to play music, such as a specific song, album, playlist, etc. In this example, the keywords may be one or words identifying one or more zones in which the music is to be played, such as the Living Room and the Dining Room shown in Figure 1 A. In some examples, the voice utterance portion 328b can include other information, such as detected pauses (e.g., periods of non-speech) between words spoken by a user, as shown in Figure 3F. The pauses may demarcate the locations of separate commands, keywords, or other information spoke by the user within the voice utterance portion 328b.

[OHl] In some embodiments, the media playback system 100 is configured to temporarily reduce the volume of audio content that it is playing while detecting the activation word portion 328a. The media playback system 100 may restore the volume after processing the voice input 328, as shown in Figure 3F. Such a process can be referred to as ducking, examples of which are disclosed in U.S. Patent No. 10,499,146, which is incorporated by reference herein in its entirety.

[0112] Figures 4A-4D are schematic diagrams of a control device 430 (e.g., the control device 130a of Figure 1H, a smartphone, a tablet, a dedicated control device, an loT device, and/or another suitable device) showing corresponding user interface displays in various states of operation. A first user interface display 431a (Figure 4A) includes a display name 433a (i.e., “Rooms”). A selected group region 433b displays audio content information (e.g., artist name, track name, album art) of audio content played back in the selected group and/or zone. Group regions 433c and 433d display corresponding group and/or zone name, and audio content information audio content played back or next in a playback queue of the respective group or zone. An audio content region 433e includes information related to audio content in the selected group and/or zone (i.e., the group and/or zone indicated in the selected group region 433b). A lower display region 433f is configured to receive touch input to display one or more other user interface displays. For example, if a user selects “Browse” in the lower display region 433f, the control device 430 can be configured to output a second user interface display 43 lb (Figure 4B) comprising a plurality of music services 433g (e.g., Spotify, Radio by Tunein, Apple Music, Pandora, Amazon, TV, local music, line-in) through which the user can browse and from which the user can select media content for play back via one or more playback devices (e.g., one of the playback devices 110 of Figure 1A). Alternatively, if the user selects “My Sonos” in the lower display region 433f, the control device 430 can be configured to output a third user interface display 431c (Figure 4C). A first media content region 433h can include graphical representations (e.g., album art) corresponding to individual albums, stations, or playlists. A second media content region 433i can include graphical representations (e.g., album art) corresponding to individual songs, tracks, or other media content. If the user selects a graphical representation 433j (Figure 4C), the control device 430 can be configured to begin play back of audio content corresponding to the graphical representation 433j and output a fourth user interface display 43 Id that includes an enlarged version of the graphical representation 433j , media content information 433k (e.g., track name, artist, album), transport controls 433m (e.g., play, previous, next, pause, volume), and indication 433n of the currently selected group and/or zone name.

[0113] Figure 5 is a message flow diagram illustrating data exchanges between devices of the media playback system 100 (Figures 1A-1M). [0114] At step 550a, the media playback system 100 receives an indication of selected media content (e.g., one or more songs, albums, playlists, podcasts, videos, stations) via the control device 130a. The selected media content can comprise, for example, media items stored locally on or more devices (e.g., the audio source 105 of Figure 1C) connected to the media playback system and/or media items stored on one or more media service servers (one or more of the remote computing devices 106 of Figure IB). In response to receiving the indication of the selected media content, the control device 130a transmits a message 551a to the playback device 110a (Figures 1A-1C) to add the selected media content to a playback queue on the playback device 110a.

[0115] At step 550b, the playback device 110a receives the message 551a and adds the selected media content to the playback queue for play back.

[0116] At step 550c, the control device 130a receives input corresponding to a command to play back the selected media content. In response to receiving the input corresponding to the command to play back the selected media content, the control device 130a transmits a message 551b to the playback device 110a causing the playback device 110a to play back the selected media content. In response to receiving the message 551b, the playback device 110a transmits a message 551c to the computing device 106a requesting the selected media content. The computing device 106a, in response to receiving the message 551c, transmits a message 55 Id comprising data (e.g., audio data, video data, a URL, a URI) corresponding to the requested media content.

[0117] At step 550d, the playback device 110a receives the message 55 Id with the data corresponding to the requested media content and plays back the associated media content.

[0118] At step 550e, the playback device 110a optionally causes one or more other devices to play back the selected media content. In one example, the playback device 110a is one of a bonded zone of two or more players (Figure IM). The playback device 110a can receive the selected media content and transmit all or a portion of the media content to other devices in the bonded zone. In another example, the playback device 110a is a coordinator of a group and is configured to transmit and receive timing information from one or more other devices in the group. The other one or more devices in the group can receive the selected media content from the computing device 106a, and begin playback of the selected media content in response to a message from the playback device 110a such that all of the devices in the group play back the selected media content in synchrony. IV. Positioning System Examples

[0119] As discussed above, a plurality of playback devices 110 and/or NMDs 120 can be distributed within an environment 101, such as a user’s home, or a commercial space such as a restaurant, retail store, mall, hotel, etc. Some of the devices may be in relatively fixed locations within the environment 101, whereas others may be portable and be frequently moved from one location to another. According to certain aspects, a positioning system can be implemented to determine relative positioning of devices within the environment 101 and optionally to control or modify behavior of one or more devices based on the relative positions. Positioning or localization information can be acquired through various techniques, optionally using sensors in some instances, examples of which are discussed below. In certain examples, one or more devices in the MPS 100, such as one or more playback devices 110, NMDs 120, or controller devices 130 may host a localization application that may implement operations (also referred to herein as functional capabilities or functionalities) that process localization information to enhance user experiences with the MPS 100. Examples of such operations include sophisticated acoustic manipulation (e.g., functional capabilities directed to psychoacoustic effects during audio playback) and autonomous device configuration and/or reconfiguration (e.g., functional capabilities directed to detection and configuration of new devices or devices that have moved or otherwise been changed in some way), among others. The requirements that these operations place on localization information vary, with some operations requiring low latency, high precision localization information and other operations being able to operate using high latency, low precision localization information.

[0120] According to certain examples, a positioning system can be implemented in the MPS 100 using a variety of different devices to generate the localization information utilized by certain application functionalities. However, the number, arrangement, and configuration of these devices can vary between examples. Additionally, or alternatively, the communications technology and/or sensors employed by the devices can vary. Given the number of variables in play within any particular MPS and the concomitant inefficiencies that this variability imposes on MPS application operation development and maintenance, some examples disclosed herein utilize one or more playback devices 110, NMDs 120, or controller devices 130 to implement a positioning system using a common positioning application programming interface (API) that decouples the positioning/localization information from specific devices or underlying enabling technologies, as illustrated conceptually in Figure 6.

[0121] Referring to Figure 6, any one or more playback devices 110, NMDs 120, or controller devices 130 in the MPS 100 (“MPS devices”) can host a positioning system application 600. In certain implementations, one or more remote computing devices can facilitate hosting the application. The positioning system application 600 implements an application programming interface (API) that exposes positioning/localization information, and metadata pertinent thereto, to MPS application functionalities 602. The MPS functionalities 602 may include a wide variety of functional capabilities relating to various user experiences and aspects of the operation of the MPS 100. For example, the MPS functionalities 602 may include one or more VAS capabilities 604, such as voice disambiguation capabilities and arbitration between different NMDs receiving the same voice inputs, for example. The MPS functionalities 602 may also include one or more MPS and/or device configuration capabilities 606, such as automatic home theater configuration or reconfiguration, dynamically accommodating portable playback devices in home theater environments, dynamic room assignment for portable playback devices or their associated docks, and contextual orientation of controller devices 130, to name a few. The MPS functionalities 602 may further include one or more other functional capabilities 608 that use positioning/localization information. To support these and other MPS functionalities 602, positioning/localization information may be used to determine various pieces of information related to the locations of MPS devices within the environment 101. For example, the positioning/localization information may be used by some MPS functionalities 602 to keep track of which playback devices 110 or NMDs 120 are in a given room or space (e.g., which playback devices are in the Living Room 10 If, in which room is playback device HOd, or which playback devices 110 are closest to the controller device 130). The positioning/localization information may further be used to determine the distance and/or orientation between playback devices 110 (with varying levels of precision), or to determine the acoustic space around NMDs 120 or NMD-equipped playback devices 110 (e.g., which playback devices 110 can be heard from NMD 120a). Thus, the positioning/localization information may be used to determine information about the topology of the MPS 100 within the environment 101, which information may then be used to automatically and dynamically create or modify user experiences with the MPS 100 and support the MPS functionalities 602. [0122] The positioning/localization information and metadata exposed by the positioning system application 600 may vary depending on the underlying communications technologies and/or sensor capabilities 610 within the MPS devices that are used to acquire the information and/or the needs of the particular MPS functionality 602. For example, certain MPS devices may be equipped with one or more network interfaces 224 that support any one or more of the following communications capabilities: BLUETOOTH 612, WIFI 614 or ultra-wide-band technology (UWB 616; a short-range radio frequency communications technology). Further, certain MPS devices may be equipped to support signaling via acoustic signaling 618, ultrasound 620, or other signaling and/or communications means 622. Certain technologies 610 may be well-suited to certain MPS functionalities 602 while others may be more useful in other circumstances. For example, UWB 616 may provide high precision distance measurements, whereas WIFI 614 (e.g., using RSSI signal strength measurements) or ultrasound 620 may provide “room-level” topology information (e.g., presence detection indicating that a particular MPS device is within a particular room or space of the environment 101). In some examples, combinations of the different technologies 610 may be used to enhance the accuracy and/or certainty of the information derived from the positioning/localization information received from one or more MPS devices via the positioning system application 600. For example, as discussed further below, in some instances, presence detection may be performed primarily using ultrasound 620; however, RSSI measurements may be used to confirm the presence detection and/or provide more precise localization information in addition to the presence detection.

[0123] Examples of MPS devices equipped with ultrasonic presence detection are disclosed in U.S. Patent Publication Nos. 2022/0066008 and 2022/0261212, each of which is hereby incorporated herein by reference in its entirety for all purposes. Examples of localizing MPS devices based on RSSI measurements are disclosed in U.S. Patent Publication No. 2021/0099736, which is herein incorporated by reference in its entirety for all purposes. Examples of performing location estimation of MPS devices using WIFI 614 are disclosed in U.S. Patent Publication No. 2021/0297168, which is herein incorporated by reference in its entirety for all purposes.

[0124] In addition to the positioning/localization information itself, some examples of the positioning system application 600 can expose metadata that specifies localization capabilities of the host MPS device, such as precision and latency information and availability of the various underlying capabilities 610. As such, the positioning system application 600 enables the MPS functionalities 602 each to utilize a common set of API calls to identify the localization capability present within their host MPS device and to access positioning/localization information made available through the identified capabilities 610.

[0125] As shown in Figure 6 and discussed above, the positioning system application 600 can interoperate with MPS devices that support a wide variety of localization capabilities, such as BLUETOOTH 612, WI-FI 614, UWB 616, acoustic signaling 618 and/or ultrasound 620, among others 622. In some examples, the positioning system application 600 includes one or more adapters configured to communicate with MPS devices using syntax and semantics specific to the localization capability 610 of the MPS devices. This architecture shields the MPS functionalities 602 from the complexity of interoperating with each type of MPS device. In some examples, each adapter can receive and process a stream of positioning/localization data from the MPS devices using any one or more of the communications capabilities 610. The adapters can interoperate with an accumulation engine within the positioning system application 600 that analyzes and merges (e.g., using a set of configurable rules) positioning/localization data obtained by the adapters and populates data structures that contain the positioning/localization information and the metadata described above. These data structures, in turn, are accessed and the positioning/localization information, and metadata, are retrieved by the positioning system application 600 in response to API calls received by the positioning system application 600 to support the MPS functionalities 602. The positioning/localization information, and metadata, can specify, in some examples, position/location of a device relative to other device, absolute position/location (e.g., within a coordinate system) of a device, presence of device (e.g., within a structure, room, or as a simple Boolean value), and/or orientation of a device.

[0126] For instance, in some examples, the positioning/localization information is expressed in two dimensions (e.g., as coordinates in a Cartesian plane), in three dimensions (e.g., as coordinates in a Cartesian space), or as coordinates within other coordinate systems. In certain examples, the positioning/localization information is stored in one or more data structures that include one or more records of fields typed and allocated to store portions of the information. For instance, in at least one example, the records are configured to store timestamps in association with values indicative of location coordinates of a portable playback device taken at a time given by the associated timestamp. Further, in at least one example, the records are configured to store timestamps in association with values indicative of a velocity of a portable playback device taken at a time given by the associated timestamp. Further, in at least one example, the records are configured to store timestamps in association with values indicative of a segment of movement (starting and ending coordinates) of a portable playback device taken at times given by associated timestamps. Other examples of positioning/localization information, and structures configured to store the same, will be apparent in view of this disclosure.

[0127] It should be noted that the API and adapters implemented by the positioning system application 600 may adhere to a variety of architectural styles and interoperability standards. For instance, in one example, the API is a web services interface implemented using a representational state transfer (REST) architectural style. In this example, the API communications are encoded in Hypertext Transfer Protocol (HTTP) along with JavaScript Object Notation and/or extensible markup language. In some examples, portions of the HTTP communications are encrypted to increase security. Alternatively, or additionally, in some examples, the API is implemented as a .NET web API that responds to HTTP posts to particular URLs (API endpoints) with localization data or metadata. Alternatively, or additionally, in some examples, the API is implemented using simple file transfer protocol commands. Also, in some examples, the adapters are implemented using a proprietary application protocol accessible via a user datagram protocol socket. Thus, the adapters and the API as described herein are not limited to any particular implementation.

V. Examples of Personalization Techniques

[0128] Aspects and embodiments are directed to personalization techniques within a media playback system that may enhance user experiences, increase user awareness of new and existing functional capabilities within the media playback system, and/or encourage user involvement with their media playback system. Techniques disclosed herein may collect household pattern data (e.g., device configuration settings, such as volume, playlist selection, etc., device movement within the environment, bonding information, etc.) to use in predicting user preferences while also taking steps to maintain user privacy, as discussed further below. In addition, personalization techniques disclosed herein may include aspects to address uncertainty when determining whether or in what manner to execute system personalization actions and/or train the personalization models to minimize friction for user adoption and exploration.

[0129] Routines are an important aspect of how users interact with technology, and in particular, of how users engage with audio and music through their media playback systems. Users typically listen differently at different times of day; for example, from some quiet background music to help them focus through the workday, to creating a party atmosphere when having friends over in the evening. While generalized trends and shifts in routine can provide useful information, techniques disclosed herein provide additional value through the ability to adapt personalization to behaviors of individual users. For example, one user may consistently choose a certain type of playlist or radio station and set the volume fairly high on a playback device in one room each morning, possibly indicating a workout routine, while otherwise having their playback devices inactive during the day and selecting a lower volume setting for some time during the evening. Another user might consistently have the volume on one or more players set relatively low throughout the day, while having it slightly louder around early evening. Listening behaviors can vary significantly among different users and therefore there can be value in user-specific personalization, rather than relying on generalized rulebased configurations.

[0130] Various recommendation systems exist that provide an approach for some degree of personalization. These recommendation systems generally offer users suggestions from a set of items based on some other item that was previously selected by the user. For example, a streaming service may recommend program D if a user has previously watched programs A, B, and C because collected data indicates that users who have watched programs A, B, and C, typically also watch program D. This approach relies on collecting vast quantities of data from a large population of users. In contrast to this list-based approach, techniques disclosed herein monitor specific user interactions with that user’s media playback system to detect individual patterns of behavior and offer or apply personalization settings unique to that user based on the detected patterns. In this context, a user can be an individual person or a group of persons (e.g., a household) associated with a particular media playback system. According to certain examples, rather than identifying latent correlations (as is done in the recommendation system approach discussed above), personalization techniques disclosed herein determine when a trend or pattern within a particular media playback system has been established, such that there exists a relatively high likelihood that the user would want system configurations or behavior to be automated in the future according to this pattern.

[0131] Unlike traditional recommendation systems, certain examples involve context awareness as an important aspect in determining and applying personalization models. As discussed above, routines play a significant role in users’ interactions with their media playback system, and these routines can shift over time. For example, users may have a different routine during the week versus over the weekend, during the summer versus during the winter, or during school vacation periods versus during school semesters. As discussed further below, aspects and examples disclosed herein incorporate contextual influence in the system’s predictions, allowing the system to adapt to changing behavior over both long and short timeframes. In particular, as discussed above, examples apply continual learning and confidence indicators to robustly determine patterns and apply personalization only in high confidence scenarios, thereby reducing the likelihood of suggesting personalization settings that are undesirable.

[0132] As discussed in more detail below, one example of a personalization setting is volume personalization. There are many scenarios where the right or wrong volume can have a significant impact on user perceptions of their playback device(s) or media playback system. For example, a user may get up early in the morning, hit “play” on their playback device in the kitchen to start their morning playlist, and be unpleasantly surprised when the sound starts many decibels too high because the playback device has retained its settings from the previous evening when music was being enjoyed at a much higher volume. In such scenarios, the user may quickly hit the “volume down” button many times, trying to reduce the volume as quickly as possible. For example, the user may hit the “volume down” button 10 or even 20 times in quick succession, which is indicative of a very frustrating user experience. Aspects and examples may provide a better experience for the user by enabling the media playback system 100 to learn from user behaviors and predict based on context (e.g., time of day) and recognized behavioral patterns when the user may want room-filling sound versus a softer, more discrete volume level. For example, machine learning models can be applied to learn from user behavior in order to facilitate smarter volume (or other) interactions, thereby improving user confidence and reducing the potential for frustrating interactions.

[0133] Further examples of volume personalization and playback device grouping personalization are discussed below. However, it will be appreciated, given the benefit of this disclosure, that the personalization techniques and approaches discussed herein may be applied to a wide variety of other characteristics, configurations, and/or behaviors of one or more playback devices (or NMDs) in a media playback system.

[0134] Referring to Figure 7, there is illustrated a block diagram of one example of a personalization service that may be implemented within a media playback system, such as the media playback system 100 discussed above. The personalization service 704 receives data from a player user data source 702 and provides personalization instructions to a player 706. Based on the instructions received from the personalization service 704, the player 706 may automatically apply a personalization setting or may offer a personalization setting suggestion to a user, as discussed further below. The player 706 may be any playback device 110 or 210 discussed above, or may be any NMD 120, 320 discussed above, for example. The player user data source 702 may include any one or more playback devices 110, 210, NMDs 120, 310, or control devices 130 in the media playback system 100.

[0135] As shown in Figure 7, in one example, the personalization service 704 includes a data collector 708, a model selector 710, and a plurality of model managers 712A-N. Each model manager 712 is associated with a respective machine learning model 722. For simplicity, in Figure 7 components of only model manager 712A are shown; however, it will be appreciated all other model managers 712 include the same components. As shown, the model manager 712A includes a data ingestion engine 714, training data 716, a trainer 718, one or more sets of one or more parameters (“parameters”) 720 of the respective machine learning model 722, the respective machine learning model 722, and a recommendation engine 724. Each of these components is described further below. The personalization service 704 may be implemented, in whole or in part, on one or more network devices (e.g., playback devices 110, 210, NMDs 120, 320, or controller devices 130) within the media playback system 100, or may be implemented, in whole or in part, on a cloud network device 102, for example. The personalization service 704 may be implemented in software or using any combination of hardware and software capable of performing the functions disclosed herein.

[0136] According to certain examples, the data collector 708 collects input data from the player user data source 702. The input data collected by the data collector 708 can include any type of data representing user interactions with the media playback system 100 as well as context information (such as date, time, location, etc.) associated with the user interactions, and device configuration data (e.g., identity of playback device being affected by the user interaction, current volume level setting, whether the device is in a bonded group, and if so, with which other players, present location of the playback device, etc.). For example, the input data associated with user interactions may include volume up or down commands, a command to select a particular audio content source, such as a particular playlist, audio streaming channel, radio station, etc., a command to group or ungroup one or more playback devices, and the like. The input data may also include movement or localization information (which may represent a user’s relocation of a portable playback device from one position to another, for example) as may be obtained via the positioning system application 600 discussed above, for example. Data collection may occur at various intervals over time. For example, a data collection event may occur each time a user interacts with a network device in the media playback system or may occur at other periodic or aperiodic times. The input data collected at each data collection event is used by the personalization service 704 to learn user routines and to offer or apply personalization settings when a learned routine has been established.

[0137] According to certain examples, the model 722 is a parameterized machine learning model configured to operate based on one or more features extracted from the collected input data. Examples of features may include the time of day, the day of the week, the type of user interaction (e.g., volume up/down, play, group, etc.), and the previous/existing setting for the corresponding playback device (e.g., previous volume that the playback device was set to, or previous bonded group setting for the playback device, etc.). The respective model 722 associated with each of the model managers 712A-N may operate based on different features. For example, different models may be applied for different personalization configurations, such as a volume level model, a bonded group model, etc., and different features may be relevant for these different personalization configurations. In one example of a volume level model, a set of features that can be used to learn trends in volume interactions includes: time since the start of the day (which allows the model to learn how volume interactions change over the course of the day), time since the start of the week (which allows the model to learn how volume interactions change over the course of the week), previous volume (which may provide volume- related context for the interaction), and type of interaction (e.g., volume, volume down, or play, which allows the model to determine, based on a given current volume level, whether the volume should likely be increased, decreased, or kept the same). Similarly, sets of relevant features can be selected for other personalization models, as will be appreciated given the benefit of this disclosure.

[0138] In addition to selecting an appropriate feature set for each personalization model, different types of machine learning models can be selected for different applications. According to certain examples, it is desirable to select a machine learning model that is capable of adapting to a variety of different contexts and to shifting routines over time, that can accommodate uncertainty (e.g., by using confidence indicators, as discussed above), and that is capable of learning based on relatively little data (e.g., hundreds to thousands of data points, rather than millions of data points). In one example, the model 722 is a Gaussian Process (GP) model. Gaussian Process models do not require large data sets, facilitate principled model uncertainty estimation, and can be tailored to specific tasks or patterns in the data through selection and/or configuration of the covariance function (kernel). Gaussian Process models can be used to interpret data with a strong periodic component (as many user behavioral patterns have) using a periodic kernel. Thus, a Gaussian Process model can be configured to encode periodic information related to user routines, which may be particularly relevant due to the strong periodicity present in many user interactions.

[0139] Examples of kernels that can be used for a Gaussian Process model 722 include a Gaussian kernel, a Matern kernel, and a periodic kernel. In addition to these kernels, white noise kernels are also used in some examples to account for variation in the input data.

[0140] The Gaussian kernel is described by the function:

In the function Fl, I represents lengthscale, which is the learned parameter of the model. Using the Gaussian kernel, the similarity between data points increases with the square of their distance. [0141] The Matern kernel is a generalization of the Gaussian kernel, allowing the smoothness of the corresponding function, F2, to be controlled via the parameter v. The additional flexibility allows the Matern kernel to adapt to “real world” data that may have significant variability. The Matern kernel is described by the function, F2:

In the function, F2, 1 represents lengthscale, which is the learned parameter of the model, and v is the smoothness parameter.

[0142] The periodic kernel is described by the function:

In the function, F3, 1 represents lengthscale and p represents periodicity, which are both learned parameters of the model. Using the periodic kernel, data points are similar if they occur in similar regions of a periodic function. For example, 7 pm on Tuesday may be similar to 6:45 pm on Wednesday.

[0143] In some examples of the model 722, multiple kernels are combined in a Gaussian Process model to produce a more expressive covariance function. In addition, as discussed further below, in some examples, multiple models 722 are combined to enhance the system’s predictive performance based on the available input data.

[0144] Still referring to Figure 7, in some examples, the data collector 708 may process the input data according to the various features associated with the models 722 to identify different player user data types and categorize the data accordingly. In addition, the data collector 708 may also tag or categorize the input data based on certain contexts or identities associated with a given data collection event. For example, the data collector 708 may categorize data received via voice commands from user A separately from data received via voice commands from user B, so as to allow the system to learn different patterns and personalization predictions for the two individual users. In another example, input data associated with a particular playback device or group of playback devices can be tagged to be associated with that particular playback device or group of playback devices. This may allow the system to learn different patterns regarding the same feature (e.g., volume personalization) that may apply to different playback devices. For example, a user may consistently choose certain volume settings when using the playback device 11 Of in the office 101 e and consistently choose different volume settings when using the playback device 110c on the patio lOli. Player user data types can also be based on the type of command or activity detected, for example a volume level data type, a bonding group data type, an audio content selection data type, etc. Thus, for each data collection event, the data collection may categorize the corresponding input data into one or more player user data types. The data collector may also apply a time stamp to each collected player user data type since, as discussed above, many potential behavioral patterns have a time component. Accordingly, time information may be important for the system to correctly determine behavioral patterns and trends. The time stamp may include time of day as well as date information.

[0145] Once input data has been collected by the data collector 708 during a data collection event, the personalization service 704 can provide the input data, as processed by the data collector 708, to one or more of the model managers 712A-N to be used to train the respective model 722 and/or to generate a personalization recommendation. As discussed above, different model managers 712 (with their respective models 722) can be used for different personalization settings, such as volume personalization or bonding personalization, for example. Accordingly, a given sample of input data acquired during a given data collection event may be relevant to one or more of the model managers 712, but potentially not to others. For example, input data corresponding to a command to group two playback devices and localization information for the two playback devices is relevant to a bonding personalization model, but may not be relevant to a volume personalization model. Accordingly, the model selector 710 can evaluate input data samples acquired by the data collector 708 and direct the input data samples to the appropriate model managers 712.

[0146] Referring to Figure 8, there is illustrated a block diagram of one example of a process 800 that may be performed by the model selector 710 based on an input data sample obtained by the data collector 708 at a particular data collection event. At 802, the model selector 710 identifies the player user data type(s), derived from the raw collected input data as discussed above, that are available from the data collector 708 for the current data collection event. In examples, the data collector 708 may extract relevant features and information from the raw input data acquired at the player user data source and format the input data for access by the model selector 710. In some examples, some data formatting may be performed by the player user data source 702 prior to the data being acquired by the data collector 708. In one example, available data types may be presented as tuples derived from the raw input data, such as (Pl, volume level 1, Tl), (Pl, volume level 2, Tl), indicating that the user changed the volume level from 1 to 2 on playback device Pl at time Tl. As another example, (Pl, P2, bonded group 1, Tl) may indicate that the user caused playback devices Pl and P2 to form a bonded group at time Tl. Many other variations will be apparent given the benefit of this disclosure. [0147] At 804, the model selector 710 identifies one or more machine learning models that are applicable to the available player user data types identified for the current data collection event. For example, if the input data sample includes a volume player user data type, the model selector 710 may identify the volume personalization model manager as an appropriate source for that input data sample. Similarly, the model selector 710 may pair input data samples having other user data types with other models, as appropriate.

[0148] At 806, the model selector 710 activates the one or more applicable models identified at 804. This may include passing the input data sample to the data ingestion engine 714 (Figure 7) of one or more model managers 712 associated with the one or more identified applicable models. In some examples, activating the model(s) at 806 includes requesting permission from the user to collect data for potential personalization. This request may not be performed at each data collection event, but rather may be performed when personalization is first activated in the media playback system 100, and optionally at various times thereafter. For example, after a significant time period has passed (e.g., months or more than one year), the system may confirm whether the user still wishes to permit data collection for personalization. In another example, if a user repeatedly declines personalization suggestions offered by the system, after a certain time frame or number of declines, the system at 806 may ask the user if the user still wants the system to collect data for personalization purposes.

[0149] In some examples in which the personalization service 704 is implemented at least in part in a cloud network 102, the data collector 708 may include a privacy filter that restricts certain data from being sent to the cloud network 102. In one example, at 806, the model selector 710 requests permission from the user to upload collected data, or certain types of data (e.g., data associating a user interaction with a particular individual user) to the cloud network 102. Based on user permissions obtained at 806, the privacy filter in the data collector 708 can be configured appropriately. In another example, once a given model (e.g., a volume personalization model or bonding personalization model) is deemed sufficiently trained by the personalization service 704, the data collector 708 can be configured, via the privacy filter, to stop uploading player user data types specific to that model until such time as the personalization service 704 determines that new training is needed (e.g., the system determines that the associated user routine has shifted over time such that the predictions are no longer sufficiently accurate). At that time, in some examples, the model selector 710, at 806, may request permission from the user to begin uploading the relevant data to the cloud network 102. [0150] Once one or more models are activated, the personalization service 704 may apply collected input data samples to those models, as appropriate based on the player user data types associated with each data sample, and run the models to train them (continuous learning) and/or to generate personalization recommendations to be acted on by the player 706.

[0151] Referring to Figure 9 (and with continuing reference to Figure 7), there is illustrated an example of a process 900 that may be implemented by a model manager 712 (e.g., the model manager 712A) to train the respective model 722 and produce a personalization recommendation.

[0152] At 902, the data ingestion engine 714 receives a data sample. As discussed above, the data sample may be obtained by the data collector 708 from the player user data source 702, optionally processed/formatted by the data collector 708, and directed to the model manager 712A by the model selector 710.

[0153] At 904, the data ingestion engine 714 processes the data sample to derive one or more features corresponding to the model 722. As discussed above, each model 722 may operate based on a selected set of features. For example, a volume level personalization model may operate on a set of features that includes: time since the start of the day, time since the start of the week, previous volume setting for the identified playback device(s), and type of interaction (e.g., volume up, volume down, or play). Accordingly, the data ingestion engine 714 processes the data sample obtained at 902 to derive these features that can be input to the model 722.

[0154] In one example, the data ingestion engine 714 transforms the sample features derived from the data sample into a format appropriate for use by the model 722 depending on the configuration of the model parameters. For example, as discussed above, a data sample may be timestamped, in the form of a Universal Standard Time (UTC) timestamp, for example, by the data collector 708, or in some examples, by the player user data source 702. The data ingestion engine 714 may process the timestamped data sample to extract /derive the feature of the time since start of the day. In some examples, the data ingestion engine 714 further processes the timestamp into floating point values to quantize the information into certain periods, such as tenths of an hour (6 minute blocks), for example. In such an example, the time 18:30 is represented as 18.5. The feature of the time since start of the week can extracted in a similar manner, but by processing the timestamp to find the time since the start of the week rather than since the start of the day. In one example, the feature of the previous volume setting may be extracted as a value between 1 and 10, for example or between 0 and 100, or another range of values depending on the model configuration. In one example, the feature of the type of user interaction (action type) may be extracted as an integer value. For example, as discussed above, a volume personalization model may consider user interactions corresponding to “Play,” “Volume Up,” and “Volume Down.” In one example, these interactions are converted to integer values as follows: Volume Down = 0, Volume Up = 1, and Play = 2. However, any of many other representations for the action type feature may be implemented in other examples. In some examples, all feature input values are scaled to zero mean and unit variance prior to being passed from the data ingestion engine 714 to the model 722.

[0155] At 906, the data ingestion engine 714 determines whether or not the data sample obtained at 902 corresponds to training data for the model 722. In some examples, for each model 722, the personalization service 704 may accumulate training data for a certain amount of time before the model 722 is run in its predictive mode to generate personalization recommendations. For example, when a given model 722 is first activated, the personalization service 704 may collect data samples over the course of a week or two to accumulate training data for that model 722. Once the model 722 is sufficiently trained, the model can begin to produce “live” personalization recommendations that are passed to the recommendation engine 724, as discussed further below. As discussed above, in certain examples, the model 722 can be configured to undergo continuous learning. Thus, in such examples, a given data sample may be both identified as training data at 906 and also used by the model 722 in its predictive mode.

[0156] If the data sample is identified at 906 as training data, the features extracted from the data sample at 904 are stored in the training data set 716 at 908. The training data set 716 accumulated at 908 is used to train the model 722 in a training mode operated by the trainer 718. In some examples, the model manager 712A can be configured such that training occurs periodically (e.g., nightly) to ensure that the model 722 is updated with respect to recent usage patterns.

[0157] Performance of the model 722 may vary according to the amount and quality of the data in the training data set 716. A consideration for a personalization model is the amount of training data required for the model 722 to generate good predictions. The number of data points and the duration of time covered by the data points both can be factors influencing the quality of the training data set 716. In some instances, periodicity of the behavior associated with the patterns being learned by the model 722 drives the amount and/or type of training data needed. For example, a training data set that includes 200 data points all from between 3pm and 6pm on various Fridays may be far less useful than a training data set that includes 100 data points capturing information about system usage over the full week. This is because many users have weekly routines, and data from only Fridays may therefore provide far less insight into the user’s routines than data covering all days of the week. [0158] Figure 10 is a graph illustrating an example of a volume data set 1000 collected for a household. As can be seen, while there is significant variation in the volume settings covered by the data set 1000, fairly strong periodicity can also be identified. A data set such as the volume data set 1000 can be used to train a volume personalization model to predict volume settings for the corresponding playback device(s) over the course of the day and week. Specifically, by identifying the periodicity, and therefore trends, present in the training data set, the model 722 can learn to predict what volume setting the user would likely want at a given time on a given day. As discussed above, periodicity in the data can also influence the choice of kernel for a given model 722.

[0159] In some examples, for volume personalization based on the four features discussed above, including action type, a Matern kernel for a Gaussian Process model 722 may provide a good solution. In some instances, however it may be desirable to train the model 722 to perform volume level predictions without associated user interactions (referred to herein as “action agnostic volume prediction”). For example, there may be instances where it would be advantageous to allow the player 706 to set the volume level ahead of the user initiating an interaction as this could ensure that the volume is set correctly before the user begins interacting with the player 706. In such examples, recognizing the periodicity (temporal features) in the training data set 716 (e.g., the volume data set 1000 or the like) may be of particular importance. Accordingly, using a periodic kernel in a Gaussian Process model 722 may provide good performance. In some examples, a combined Matern and periodic kernel can be used.

[0160] Returning to Figure 9, at 914 the trainer 718 evaluates the training data set 716, including the sample features from the data sample obtained at 902, to determine whether change criteria for the training data set 716 have been met. Change criteria may be met if any features added to the training data set 716 at 908 are new or sufficiently different from existing features collected in the training data set 716 to warrant updating of the model 722.

[0161] If the change criteria are met, then at 916, the trainer 718 uses the training data set 716 to re-trains the model and, thereby, update one or more model parameters 720. For example, the trainer 718 may execute a gradient-based optimizer to adjust the lengthscale parameter of a model within operation 916.

[0162] Once trained, at 910, the model 722 generates one or more personalization recommendation(s) and associated confidence metric(s). The sample features derived from the data sample are passed to the model 722 by the data ingestion engine 714. The model 722 uses the sample features to generate the one or more personalization recommendation(s) and associated confidence metric(s). As discussed above, in certain examples (for example, where the model 722 is a Gaussian Process model), the model 722 is configured to provide principled model uncertainty estimates. This means that the personalization service 704 can have guidance as to whether the corresponding recommendation generated by the model 722 is likely to be correct, because the uncertainty estimates may convey information about a lack of data (e.g. for times when the user does not use a playback device) and information about the variance in user activity. For example, referring again to the graph illustrated in Figure 10, from the observed values, it can be seen that certain time periods are more consistent, such as late afternoon/evening, whereas others have a wider variance, such as late evening/night. Uncertainty estimates can capture this variance, enabling the model 722 to tag or associate a volume prediction for a time during the late evening/night with a lower confidence metric than a volume prediction for a time during the late afternoon/evening, for example. Confidence metrics may be used to determine actions taken by the recommendation engine 724, as discussed further below. In particular, confidence metrics provide a valuable resource in terms of configuring the personalization service 704 to provide useful personalization recommendations to users and reduce instances of providing incorrect, unwanted, or annoying personalization suggestions or actions.

[0163] In certain examples, the model 722 outputs one or more recommendations, along with associated confidence metric(s), to the recommendation engine 724. At 912, the recommendation engine 724 may then execute a player operation based on the recommendation(s) and confidence metric(s). Executing a player operation at 912 may include instructing the player 706 to automatically take an action (e.g., set a volume level, group or ungroup with one or more other playback devices, begin playback of certain audio content, etc.), or may include directing the player 706 to offer the user a recommendation, as discussed further below.

[0164] In some instances, the recommendation engine 724 executes the player operation in direct or immediate response to receiving the recommendation from the model 722. In another example, the recommendation engine executes the player operation at certain time points or time intervals. For example, in the context of volume personalization, the recommendation engine 724 may receive from the model 722 predictions (recommendations) for the volume setting for the player 706 at different times throughout the day (e.g., hourly predictions). While the player 706 is inactive, the recommendation engine 724 may periodically (e.g., hourly) update the suggested volume setting for the player 706 based on the periodic predictions provided by the model 722. In some instances, as in the case of action agnostic volume prediction discussed above, for example, the recommendation engine 724 may automatically instruct the player 706 (while the player 706 is inactive) to set its volume level based on the periodic predictions, such that when a user does interact with the player 706, the volume is set to the anticipated correct level.

[0165] In examples, the system may monitor the user’s response to such automatic volume settings to gauge the accuracy of the prediction. For example, if a user activates the player 706 and does not adjust the volume level, the system may interpret that the predicted volume level setting was correct. On the other hand, if the user immediately changes the volume, the system may interpret that the volume prediction was not correct. In certain examples, this user feedback can be incorporated into the training data set 716 as labeled features that weight confirmed/rejected settings more heavily. For example, the positive/negative user feedback can be associated with the volume setting for that particular player 706 for the time period at which the user interaction occurred, and increase the weight of that setting at that time period. By retraining the model 722 with this labeled training data, the model 722 may produce similar recommended settings with higher or lower confidence metrics. In some examples, where the labeled training data is based on positive user feedback, the re-trained model may produce a corresponding volume prediction with a higher confidence metric. In other examples, where the labeled training data is based on negative user feedback, the model may be less likely to produce a corresponding volume prediction, or if it does, may produce the corresponding volume prediction with a lower confidence metric. In some instances, weighting may depend on whether the user alters the volume setting by only a little (indicating that the prediction was close, but not perfect) or lot (indicating that the prediction was incorrect). Thus, the stronger the confirmation/rej ection of the recommendation, the heavier the weight may be that is placed on the associated labeled features in the training data.

[0166] According to certain examples, if the confidence metric associated with a given recommendation indicates that the system has high confidence that the prediction is correct (e.g., the confidence metric is above some predefined threshold), the recommendation engine 724 may instruct the player 706 to automatically apply the recommended personalization setting. For example, the player 706 can be directed to automatically set its volume level, as discussed above.

[0167] If the confidence metric for a given recommendation has a mid-range value (e.g., below the predefined threshold indicating “high” confidence, but above another threshold indicating “low” confidence), the recommendation engine 724 may direct the player 706 to offer the recommendation to the user and/or ask for confirmation from the user as to whether or not to apply the personalization recommendation. As discussed above, the user feedback may be incorporated by the model 722 to impact future recommendations. In some examples, the player 706 may directly ask the user for feedback using a voice output (e.g., produced from text-to-voice using voice/speech conversion components 312o (Figure 3E) via one or more of its transducers 214, 314 (e.g., Figures 2B, 3C). In another example, the recommendation engine 724 may direct the player 706 to provide a visual indication of an available recommendation, for example, by illuminating, changing the color of, and/or flashing one or more visual indicators (e.g., indicators 313e and/or 313f) on the player’s user interface. Upon noticing the visual indication, the user may interact with the player 706 to find out about the available personalization recommendation(s) and provide user feedback, positive or negative, in response. In this manner, the personalization service 704 may encourage user engagement with their playback devices and exploration of available options in a non-intrusive or non-disruptive way.

[0168] In some examples, if the confidence metric indicates that the confidence level for the associated recommendation is low (e.g., the confidence metric is below the predefined threshold indicating low confidence), the recommendation engine 724 may not execute a player operation at 912. Instead, the system may continue to monitor and collect data regarding the personalization attribute and add to the training data set 716 until the model 722 is sufficiently trained to produce a recommendation with a medium (mid-range) or high confidence metric. [0169] Referring to Figures 7 and 9, as discussed above, different model kernels can provide better performance under different circumstances and scenarios. In addition to combining kernels, as discussed above, according to certain embodiments, dynamic model switching can be used to allow an appropriate model to be applied (at 910) in a given circumstance by dynamically selecting one or more model managers 712, and training the respective model 722, according to the available input data. In examples, an ensemble of multiple model managers 712 potentially may be used for a particular personalization application (e.g., volume personalization or grouping personalization), with the respective model 722 associated with each of the multiple model managers 712 implementing a different kernel (or combination of kernels). This concept is based on the principle that, depending on the available input data and/or training data, some models may be more preferable (more likely to produce good predictions) than others.

[0170] Different models can be used in different contexts. For example, in one instance of the case of action agnostic volume prediction, the volume level of the player 706 may be preemptively set at certain intervals (e.g., hourly) according to learned user trends. In another example of action agnostic volume prediction, the volume level of the player 706 can be automatically set, according to learned user trends, when the user initiates playback of audio content on the player 706. This approach facilitates greater temporal sensitivity than the period preemptive approach. As discussed above, in either case, a Gaussian Process model with a combined periodic and Matem kernel may be a good model choice for such applications. In another example, the system can be configured to implement action-aware interaction-based volume setting. In this case, the volume level of the player 706 is set when the user interacts with the player 706, using information about whether the interaction is a “Play,” “Volume Up,” or “Volume Down” interaction. In such examples, a Gaussian Process model with a Matern kernel may be a preferred model choice. While an action-aware approach may yield more consistently accurate results, this approach requires volume interaction information, which may not always be available to the personalization service 704. Thus, in some circumstances, it may be preferable to use a model with one kernel and in other circumstances, a model with a different kernel may be preferred. Accordingly, dynamically switching between models allows for the most appropriate model to be used based on the available data.

[0171] Dynamic model switching may be applied according to several different methods or approaches, and may be implemented by the model selector 710. Certain examples apply uncertainty-based model switching. In this case, the model manager 712 (and its associated model 722) is selected according to the prediction with the lowest uncertainty. Thus, for example, the model selector may pass the data sample to two or more model managers 712, and each respective model 722 may produce, at 910, a personalization recommendation and an associated confidence metric. The recommendation engines 724 may provide the results to the model selector 710, which may instruct the recommendation engine 724 having the result with the higher confidence metric to execute the player operation at 912. In some examples, the model selector may retain knowledge of the results, such that in future similar circumstances, the data sample may be provided only to the model manager 712 having the model 722 that produced a better earlier result. This approach is based on the assumption that Gaussian Process uncertainties should be well calibrated, and therefore the model that produces a recommendation with a higher confidence metric is more likely to produce the correct prediction.

[0172] According to another example, the personalization service 704 can be configured to implement action-dependent model switching. In this case, the model manager 712 (with its respective model 722) is selected for training based on the available input data. For example, in the volume personalization example discussed above, if action type feature information is available, the model selector 710 may select a model manager 712 with a model 722 that uses a Matem kernel, whereas if the action type feature information is not available, the model selector 710 may select a model manager 712 with a model 722 that uses a combined Matern and periodic kernel. Numerous other variations will be apparent given the benefit of this disclosure.

[0173] Referring to Figure 11, there is illustrated a sequence diagram for one example of a process of applying the personalization service for volume level personalization in accord with certain aspects. As discussed above, the data collector 708 collects input data that contains various player user data types associated with volume personalization. In the illustrated example, a data sample 1102 provided from the data collector 708 to the data ingestion engine 714 includes a “Play” command. As discussed above, the data ingestion engine 714 processes the data sample 1102 to extract one or more volume personalization features 1104. In this example, the volume features 1104 include an action type feature (i.e., the user interaction associated with the data sample 1102, which in this example is a “Play” command). The data ingestion engine 714 passes the extracted volume features to a selected trained model 722. In this example, the volume features include an action type feature, therefore, as discussed above, in some examples, the model 722 may be a Gaussian Process model with a Matern kernel. The model 722 operates on the volume features 1104 to provide an output 1106 that includes a recommended volume level and a confidence metric associated with the recommended volume level. The recommendation engine 724 takes the output 1106 from the model 722 and executes a player operation, as discussed above. In one example, the recommendation engine 724 provides a recommended volume action 1108 to the player 706. As discussed above, the recommended volume action 1108 may take any of various forms. For example, the recommended volume action 1108 may be an instruction to automatically set the volume level of the player 706 to the recommended volume level. In another example, the recommended volume action 1108 may be an instruction to direct the player 706 to offer the recommendation to the user, either by a voice prompt or visual indicator, as discussed above. The recommended volume action 1108 may be determined, at least in part, by the confidence metric associated with the recommended volume level, as also discussed above.

[0174] It will be appreciated, given the benefit of this disclosure, that the personalization service 704 can be implemented in a variety of ways and can incorporate various machine learning approaches. Figure 16 illustrates another example of an implementation of at least some of the functionalities of the personalization service 704. In this example, the personalization service 702 includes a model predictive controller 1620 that may implement some or all of the functionality of the model manager 712 described above, optionally in conjunction with other functionality. The model predictive controller 1620 includes a model 1622, which may be a parameterized machine learning model, such as the model 722 discussed above. The model predictive controller 1620 receives input data 1610, and the model 722 may be configured to operate based on one or more features extracted from the collected input data 1610, as described above. In some examples, the input data 1610 includes data obtained from the player user data source 702 and may include any of the types and/or forms of data described above.

[0175] According to certain examples, the model predictive controller 1620 runs the model 1622 based on parameters associated with the one or more features extracted from the input data 1610 to produce a personalization result or recommendation, as described above. Accordingly, the model predictive controller 1620 may include a data sampler 1624, which may perform the same functions described above with respect to the data collector 708 and/or the data ingestion engine 714. The model predictive controller may further acquire and store user preference information, as indicated at 1626. The user preference information may include user-provided information regarding the level of personalization desired by the user, playback device attributes or configurations that the user does or does not want to be personalized (e.g., a user may agree to volume personalization but not grouping personalization), and/or other user preferences with respect to the personalization functionalities described herein. The user preference information may be acquired as part of the input data 1610 in some examples or may be separately acquired and stored. In some examples, the model predictive controller 1620 further includes a decision module 1628, which may set one or more thresholds that determine action(s) taken by the model predictive controller in response to outputs from the model 1622. For example, as discussed above, the model 1622 may output one or more recommendations along with associated confidence metric(s). The decision module 1628 may set thresholds associated with the confidence metric(s) that determine whether the model productive controller executes a personalization action or offers a recommendation to the user, for example. In some examples, the decision module 1628 sets thresholds associated with the confidence metric(s) that establish a “trust region” in which the output from the model 1622 can be relied upon (trusted) such that the model predictive controller 1620 can act in accord with the personalization recommendation output by the model 1622. In some examples, the decision module 1628 incorporates some or all of the functionality associated with the recommendation engine 724 described above.

[0176] In the example illustrated in Figure 16, the personalization service 704 includes an optimizer 1620 that operates based on one more hyperparameters to optimize performance of the model predictive controller 1620. As described above, the model 1622 (or model 722) operates based on various parameters (variables belonging to the model). Hyperparameters are higher level variables that determine characteristics such as an architecture of the model 1622 (e.g., the choice of kernel as described above), how the model 1622 is used, variables that affect the application of the model, etc. A hyperparameter can take the form of a single continuous scalar variable or a discrete categorical variable (e.g., which kernel to use). In some examples, the optimizer 1630 selects hyperparameters to adjust the model 1622 by testing the performance of the model on a validation dataset. In some examples, the optimizer 1630 may perform some or all of the functionality described above with respect to the model selector 710 and/or the model manager 712.

[0177] Referring now to Figures 12 and 13, another example of personalization is grouping personalization, which in some instances may be based at least in part on detected movement patterns and localization information regarding one or more playback devices. In particular, grouping personalization recommendations may be provided based on recognizing and identifying repeated patterns of movement of one or more playback devices, where the pattern(s) of movement are associated with the formation (and/or dissolution) of a bonded group. An example of grouping personalization is discussed with reference to Figures 12 and 13. It will be appreciated that grouping personalization may be applied, according to the various techniques and aspects discussed above, in many other circumstances and arrangements in addition to the example shown in Figures 12 and 13.

[0178] Figure 12 illustrates a first example media playback system configuration 1200 in an environment 1202. Figure 13 illustrates a second example media playback system configuration 1300 in the environment 1202. In the configuration 1300, two playback devices Pl and P2, at least one of which may be a portable playback device, are in a bonded group, forming a bonded pair. In the configuration 1200, the playback device Pl is moved from a first room 1204 into a second room 1206, as shown in Figure 12. In one example, the movement of the playback device Pl from room 1204 into room 1206 is correlated with dissolution of the bonded pair, that is, in the configuration 1200, the playback device Pl is not grouped with the playback device P2. Over time, a consistent pattern or routine may be established in which the user 1208 moves the playback device Pl into the room 1204 and instructs the media playback system to form a bonded group including the playback devices Pl and P2, and out of the room 1204, into the room 1206, and instructs the media playback system to dissolve the group between the playback devices Pl and P2. By collecting player user data that identifies the movement between the rooms 1204 and 1206 and the associated action (grouping and ungrouping), a model 722 or 1622 can be trained to recognize the routine. Accordingly, once the model 722/1622 is trained, the personalization service 704 may predict that when the user 1208 takes the playback device Pl into the room 1204, and optionally places the playback device Pl close to the playback device P2 (which may be determined using localization information obtained via the positioning system application 600 discussed above), it is likely that the user will want to form a bonded group with the playback devices Pl and P2. Similarly, the personalization service 704 may predict that when the user 1208 moves the playback device Pl away from the playback device P2 and into the room 1206, the user will likely want to ungroup the playback devices Pl and P2. The personalization service 704 may therefore, via the recommendation engine 724 or the decision module 1628 of the model predictive controller 1620, either automatically group and/or ungroup the playback devices Pl and P2 based on the prediction or offer a recommendation to the user 1208 that the system can automatically perform this action in the future, if so desired. The user 1208 may then confirm or reject this personalization recommendation.

[0179] As discussed above, collecting the player user data that identifies the movement of the playback device Pl can be accomplished via the positioning system application 600, using any of the presence detection or location estimation techniques and technologies discussed above. In some examples, collecting movement data may include measuring a distance traveled by the playback device Pl. In some examples, determining that the playback device Pl has moved into (or out of) the rooms 1204 or 1206 can be accomplished using presence detection techniques. For example, any of playback devices P2, SI, S2, or Bl may detect the presence (or lack thereof) of the playback device Pl . In another example, playback device Pl may detect the presence (or lack thereof) of any of the playback devices P2, SI, S2, or B2. Based on known locations of the playback devices P2, SI, S2, and/or Bl, the location of the playback device Pl can be estimated using the presence detection results.

[0180] Further examples of using presence detection and other techniques to monitor the movement of a playback device are described in International Application No. PCT/US2023/075497 filed on September 39, 2023 and published under International Publication No. WO 2024/073651, which is hereby incorporated by reference herein in its entirety for all purposes.

[0181] Figure 14 illustrates another media playback system configuration 1400. In the example configuration 1400, the user 1208 has moved both playback devices Pl and P2 into the room 1206, which also contains playback devices SI, S2, and Bl. The playback devices SI, S2, and Bl may form a home theater group, for example. In the example illustrated in Figure 14, the use 1208 establishes a group set-up in which the playback devices SI and S2 are muted, and the playback devices Pl and P2 are added to a group with the playback device Bl. In some instances, as discussed above, if the configuration is sufficiently and consistently repeated, the personalization system may learn the routine automatically. However, in other instances, the user 1208 may want to accelerate the personalization process and instruct the system to record the configuration 1400 so that it can be easily reconstructed in the future without the requirement for many repetitions to train the model 722 or 1622. Accordingly, the user 1208 may issue a voice command 1210 (e.g., “Hey Sonos, remember this”), for example, (or alternatively enter a command via a control device 130) to direct the personalization service 704 to record the configuration 1400. Once recorded, when the user 1208 again positions the playback devices Pl and P2 in the locations approximately corresponding to the configuration 1400, the personalization service may automatically implement the other device settings associated with the configuration 1400 (e.g., mute playback devices SI and S2 and group the playback devices Pl and P2 with the playback device Bl) or offer the suggestion to the user 1208.

[0182] As discussed above, user feedback can be used by the personalization service 704 to train the model 722 or 1622 by increasing (based on positive user feedback) or decreasing (based on negative user feedback) weightings associated with settings in the training data set 716 corresponding to a given recommendation. In the example of Figure 14, the affirmative user feedback (e.g., provided via the voice command 1210) can be used to train the model 722/1622 by directly associating a very high confidence metric with the device settings correlated with the positional configuration of the playback devices Pl and P2 relative to the other playback devices S 1 , S2, and B 1. Thus, when the system detects a repeat of the positional configuration 1400 (as may be determined from localization data acquired via the positioning system application 600), the personalization service 704 may predict the user’s desired result based on the previously stored configuration information.

[0183] Figures 15A and 15B are sequence diagrams illustrating examples of processes that may be performed by the personalization service 704 to implement grouping personalization in accord with certain aspects.

[0184] Referring to Figure 15 A, as discussed above, the data collector 708 (or data sampler 1624) collects input data (e.g., input data 1610) that contains various player user data types associated with grouping personalization. In the illustrated example, a data sample 1502 provided from the data collector 708 to the data ingestion engine 714 (or acquired by the data sampler 1624) includes detected movement of a playback device (e.g., of the playback device Pl in the example of Figures 12 and 13). As discussed above, the data ingestion engine 714 (or the data sampler 1624) processes the data sample 1502 to extract one or more grouping personalization features 1504. In this example, the grouping personalization features 1504 may include localization features, movement features, grouping features (e.g., based on a command in the data sample 1502 to group or ungroup certain players), etc. The data ingestion engine 714 (or the data sampler 1624) passes the extracted grouping personalization features 1504 to a selected trained model 722 (or model 1622). The model 722/1622 operates on the grouping personalization features 1504 to provide an output 1506 that includes a recommended grouping behavior (e.g., bond or unbond the playback devices Pl and P2 in the example of Figures 12 and 13) and a confidence metric associated with the recommended grouping behavior. The recommendation engine 724 (or the decision module 1628) takes the output 1506 from the model 722 (or 1622) and executes a player operation, as discussed above. In one example, the recommendation engine 724 (or model decision module 1628) provides a recommended grouping action 1508 to the player 706. As discussed above, the recommended grouping action 1508 may take any of various forms. For example, the recommended grouping action 1108 may be an instruction to automatically group two or more playback devices to form a bonded group. In another example, the recommended grouping action 1508 may be an instruction to direct the player 706 (e.g., one of the playback devices Pl, P2, SI, S2, or Bl in the example of Figures 12 and 13) to offer the recommendation to the user, either by a voice prompt or visual indicator, as discussed above. The recommended grouping action 1508 may be determined, at least in part, by the confidence metric associated with the recommended grouping behavior, as also discussed above.

[0185] Referring to Figure 15B, as discussed above, in certain examples, the personalization service 704 may incorporate user feedback into the process 1500 to increase or decrease the confidence metrics and/or to train (or retrain) the model 722. For example, the data collector 708 may collect user input data (e.g., voice inputs or inputs detected via the user interface on the player user data source 702) and provide user feedback 1510 to the data ingestion engine 714. In one example, the user feedback may include a user response to automatically implemented grouping behavior or a suggestion to perform (or automate) certain grouping behavior. In another example, such as the example illustrated in Figure 14, the user feedback may be a proactive user command that is not in response to a suggestion offered or action implemented by the personalization service 704. The data ingestion engine 714 processes the user feedback 1510 to extract grouping features 1512, as discussed above. In this instance, as the grouping features 1512 include or are based on user feedback, the grouping features 1512 may be identified at 906 (Figure 9) as training data. Accordingly, the grouping features may be added to the training data set 716 and evaluated by the trainer 718, as discussed above. If the change criteria for the training data set are met, the trainer 718 adjusts the model parameters 720, as discussed above. In certain examples, such as the scenario of Figure 14, for example, the user feedback may be a heavily weighted, or even dispositive factor, in determining the parameter adjustment. For example, in the scenario of Figure 14, the user affirmatively requests the system to record the configuration and associated device settings. Therefore, in this case, the model parameters may be adjusted to heavily weight the model 722 toward generating, upon detection of that positional configuration, a grouping recommendation that corresponds to the user’s requested set-up. Similarly, where examples in which the personalization service 704 has a configuration as shown in Figure 16, the model predictive controller 1620 can also intake and apply user feedback 1510 in a manner similar to that described above with respect to the configuration of Figure 7.

[0186] Thus, examples and embodiments provide techniques for personalizing various aspects of a media playback system based on learned user routines and preferences. Through the use of personalization techniques, the user experience may be streamlined and enhanced by enabling users to achieve desired outcomes with reduced manual effort and interaction. For example, playback devices can be automatically set to predicted user-preferred volume levels, and automatic grouping behavior can be implemented or suggested, thus simplifying the actions required by the user to achieve a desired end result.

VI. Conclusion

[0187] The above discussions relating to playback devices, controller devices, playback zone configurations, and media content sources provide only some examples of operating environments within which functions and methods described below may be implemented. Other operating environments and configurations of media playback systems, playback devices, and network devices not explicitly described herein may also be applicable and suitable for implementation of the functions and methods.

[0188] The description above discloses, among other things, various example systems, methods, apparatus, and articles of manufacture including, among other components, firmware and/or software executed on hardware. It is understood that such examples are merely illustrative and should not be considered as limiting. For example, it is contemplated that any or all of the firmware, hardware, and/or software aspects or components can be embodied exclusively in hardware, exclusively in software, exclusively in firmware, or in any combination of hardware, software, and/or firmware. Accordingly, the examples provided are not the only ways to implement such systems, methods, apparatus, and/or articles of manufacture.

[0189] Additionally, references herein to “embodiment” means that a particular attribute, structure, or characteristic described in connection with the embodiment can be included in at least one example embodiment. The appearances of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. As such, the embodiments described herein, explicitly and implicitly understood by one skilled in the art, can be combined with other embodiments.

[0190] The specification is presented largely in terms of illustrative environments, systems, procedures, steps, logic blocks, processing, and other symbolic representations that directly or indirectly resemble the operations of data processing devices coupled to networks. These process descriptions and representations are typically used by those skilled in the art to most effectively convey the substance of their work to others skilled in the art. Numerous specific details are set forth to provide a thorough understanding of the present disclosure. However, it is understood to those skilled in the art that certain embodiments of the present disclosure can be practiced without certain, specific details. In other instances, well known methods, procedures, components, and circuitry have not been described in detail to avoid unnecessarily obscuring aspects of the embodiments. Accordingly, the scope of the present disclosure is defined by the appended claims rather than the foregoing description of embodiments.

[0191] When any of the appended claims are read to cover a purely software and/or firmware implementation, at least one of the elements in at least one example is hereby expressly defined to include a tangible, non-transitory medium such as a memory, DVD, CD, Blu-ray, and so on, storing the software and/or firmware.

VII. Further Examples

[0192] The following examples pertain to further embodiments, from which numerous permutations and configurations will be apparent.

[0193] Example 1 provides a method of personalizing a setting of one or more playback devices in a media playback system, the method comprising collecting, over time, sample values of the setting and feature data associated with the sample values, training a parameterized machine learning model to predict a recommended value of the setting using the sample values and the feature data, detecting input data representative of one or more of a current time, a current user interaction with the one or more playback devices, or current location of the one or more playback devices, extracting current feature data from the input data, applying the parameterized machine learning model to the current feature data to generate the recommended value of the setting and a confidence metric corresponding to the recommended value, and executing a playback device operation based on the recommendation and the confidence metric.

[0194] Example 2 includes the method of Example 1, wherein at least one element of the feature data is derived from one or more of a time at which a corresponding sample value was collected or a location of the one or more playback devices when the correspond sample value was collected.

[0195] Example 3 includes the method of one of Examples 1 or 2, wherein executing the playback device operation includes adjusting the setting to the recommended value.

[0196] Example 4 includes the method of one of Examples 1 or 2, wherein executing the playback device operation includes providing, via the one or more playback devices, one or more of an audible suggestion or a visual suggestion to adjust the setting to the recommended value.

[0197] Example 5 includes the method of Example 4, wherein providing the visual suggestion includes at least one of illuminating a light-emitting diode (LED) on at least one playback device, changing a color of light emitted by the LED on the at least one playback device, or flashing the LED on the at least one playback device.

[0198] Example 6 includes the method of any one of Examples 1-5, wherein the setting is a volume level of at least one of the one or more playback devices.

[0199] Example 7 includes the method of Example 6, wherein the current user interaction with the one or more playback devices includes one or more of a user command to adjust the volume level of the at least one playback device or a user command to begin playback of audio content. [0200] Example 8 includes the method of any one of Examples 1-5, wherein the setting is a group status of the one or more playback devices.

[0201] Example 9 includes the method of any one of Examples 1-8, wherein executing the playback device operation based on the recommendation and the confidence metric comprises one or more of adjusting the setting to the recommended value based on the confidence metric exceeding a threshold value, or causing the one or more playback devices to output a recommendation to adjust the setting to the recommended value based on the confidence metric being at or below the threshold value. [0202] Example 10 includes the method of Example 9, wherein causing the one or more playback devices to output the recommendation includes one or more of causing at least one playback device to display a visual indication, and/or causing at least one playback device to provide an audible suggestion to adjust the setting to the recommended value.

[0203] Example 11 includes the method of Example 10, wherein causing the at least one playback device to display the visual indication includes at least one of illuminating a lightemitting diode (LED) on at least one playback device, changing a color of light emitted by the LED on the at least one playback device, or flashing the LED on the at least one playback device.

[0204] Example 12 includes the method of any one of Examples 1-11, wherein the parameterized machine learning model is a Gaussian Process model.

[0205] Example 13 includes the method of any one of Examples 1-12, wherein training the parameterized machine learning model includes selecting, based on the sample values, a chosen parameterized machine learning model from among a plurality of parameterized machine learning models.

[0206] Example 14 includes a playback device configured to implement the method of any one of Examples 1-13.

[0207] Example 15 provides a network device comprising at least one processor, and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor to control the network device to implement the method of any one of Examples 1-13.

[0208] Example 16 provides a method of personalizing a volume setting of a playback device in a media playback system. The method comprises collecting, over time, sample values of the volume setting and feature data associated with the sample values, elements of the feature data being derived from one or more of a time at which a corresponding sample value was collected, and a present volume setting of the playback device when the correspond sample value was collected, training a parameterized machine learning model to predict a recommended value of the volume setting using the sample values and the feature data, detecting input data representative of one or more of a current time, and a current user interaction with the playback device, extracting current feature data from the input data, applying the parameterized machine learning model to the current feature data to generate the recommended value of the volume setting and a confidence metric corresponding to the recommended value, and executing a playback device operation based on the recommended value and the confidence metric. [0209] Example 17 includes the method of Example 16, wherein executing the playback device operation includes adjusting the volume setting of the playback device to the recommended value.

[0210] Example 18 includes the method of Example 16, wherein executing the playback device operation includes causing the playback device to output an audible suggestion to adjust the volume setting of the playback device to the recommended value.

[0211] Example 19 includes the method of any one of Examples 16-18, wherein the parameterized machine learning model is a Gaussian Process model.

[0212] Example 20 includes the method of Example 16, wherein executing the playback device operation includes providing, via the playback device, one or more of an audible suggestion or a visual suggestion to adjust the volume setting to the recommended value.

[0213] Example 21 includes the method of Example 20, wherein providing the visual suggestion includes at least one of illuminating a light-emitting diode (LED) on the playback device, changing a color of light emitted by the LED on the playback device, or flashing the LED on the playback device.

[0214] Example 22 provides a playback device configured to implement the method of any one of Examples 16-21.

[0215] Example 23 provides a method of personalizing a grouping configuration of a plurality of playback devices in a media playback system. The method comprises collecting, over time, sample values of the grouping configuration and feature data associated with the sample values, at least one element of the feature data being derived from a location of at least one of the plurality of playback devices when the correspond sample value was collected, training a parameterized machine learning model to predict a recommended grouping configuration using the sample values and the feature data, detecting input data representative of a current location of the at least one playback device, extracting current feature data from the input data, applying the parameterized machine learning model to the current feature data to generate the recommended grouping configuration and a confidence metric corresponding to the recommended grouping configuration, and executing a playback device operation based on the recommended grouping configuration and the confidence metric.

[0216] Example 24 includes the method of Example 23, wherein executing the playback device operation includes one of automatically grouping the plurality of playback devices or automatically ungrouping the at least one playback device from a remainder of the plurality of playback devices. [0217] Example 25 includes the method of Example 23, wherein executing the playback device operation includes causing the at least one playback device to output an audible suggestion to adjust a grouping status of the at least one playback device to the recommended grouping configuration.

[0218] Example 26 includes the method of any one of Examples 23-25, wherein the parameterized machine learning model is a Gaussian Process model.

[0219] Example 27 provides a playback device configured to implement the method of any one of Examples 23-26.

[0220] Example 28 provides a method of personalizing one or more settings of one or more playback devices in a media playback system, the method comprising: providing collected sample values of the one or more settings and feature data associated with the sample values to a parameterized machine learning model to train the model to predict a recommended value of the one or more settings using the sample values and the feature data; based on current feature data associated with at least one playback device, applying the parameterized machine learning model to the current feature data to generate the recommended value of the one or more settings; and causing at least one playback device to perform at least one playback device operation based on the recommended value.

[0221] Example 29 includes the method of Example 28, wherein at least one element of the feature data is derived from one or more of a time at which a corresponding sample value was collected or a location of the at least one playback device when the correspond sample value was collected.

[0222] Example 30 includes the method of one of Examples 28 or 29, wherein the current feature data is representative of one or more of a current time, a current user interaction with the one or more playback devices, or current location of the one or more playback devices.

[0223] Example 31 includes the method of any one of Examples 28-30, further comprising: collecting, over time, the sample values of the one or more settings at the media playback system, and extracting the associated feature data from the sample values.

[0224] Example 32 includes the method of Example 31, wherein collecting the sample values comprises collecting, by one or more devices of the media playback system, the sample values over time.

[0225] Example 33 includes the method of any one of Examples 28-32, further comprising detecting input data representative of the current feature data. [0226] Example 34 includes the method of any one of Examples 28-33, wherein causing the at least one playback device to perform the at least one playback device operation includes adjusting the one or more settings to the recommended value.

[0227] Example 35 includes the method of any one of Examples 28-34, wherein causing the at least one playback device to execute the at least one playback device operation includes providing, via the at least one playback device, one or more of an audible suggestion or a visual suggestion to adjust the setting to the recommended value.

[0228] Example 36 includes the method of any one of Examples 28-35, wherein the one or more settings includes a volume level of at least one of the one or more playback devices.

[0229] Example 37 includes the method of Example 30, alone or in combination with any one of Examples 28, 29, or 31-36, wherein the current user interaction with the at least one playback devices includes one or more of a user command to adjust the volume level of at least one playback device or a user command to begin playback of audio content.

[0230] Example 38 includes the method of any one of Examples 28-37, wherein the current feature data includes a group status of at least one playback device.

[0231] Example 39 includes the method of any one of Examples 28-38, wherein the one or more settings includes a grouping configuration of one or more playback devices of the media playback system.

[0232] Example 40 includes the method of Example 39, wherein the current feature data comprises a location of at least one playback device, and wherein executing the playback device operation comprises configuring a grouping configuration of a plurality of playback devices.

[0233] Example 41 includes the method of Example 40, wherein configuring the grouping configuration comprises at least one of: automatically grouping the plurality of playback devices or automatically ungrouping the at least one playback device from a remainder of the plurality of playback devices.

[0234] Example 42 includes the method of any one of Examples 28-41, wherein the parameterized machine learning model generates a confidence metric corresponding to the recommended value.

[0235] Example 43 includes the method of Example 42, wherein causing the at least one playback device to perform the at least one playback device operation based on the recommendation and the confidence metric comprises one or more of: adjusting the one or more settings to the recommended value based on the confidence metric exceeding a threshold value; or causing the at least one playback device to output a recommendation to adjust the setting to the recommended value based on the confidence metric being at or below the threshold value.

[0236] Example 44 includes the method of any one of Examples 28-43, wherein the parameterized machine learning model is a Gaussian Process model.

[0237] Example 45 includes the method of any one of Examples 28-44, wherein training the parameterized machine learning model includes selecting, based on the sample values, a chosen parameterized machine learning model from among a plurality of parameterized machine learning models.

[0238] Example 46 provides a network device comprising: at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor to control the network device to implement the method of any one of Examples 28-45.

[0239] Example 47 provides a media playback system comprising: at least one playback device; and one more devices configured for performing the method of any one of Examples 28-45.

Claims

1. A method of personalizing one or more settings of one or more playback devices in a media playback system, the method comprising: providing collected sample values of the one or more settings and feature data associated with the sample values to a parameterized machine learning model to train the model to predict a recommended value of the one or more settings using the sample values and the feature data; based on current feature data associated with at least one playback device, applying the parameterized machine learning model to the current feature data to generate the recommended value of the one or more settings; and causing at least one playback device to perform at least one playback device operation based on the recommended value.

2. The method of claim 1, wherein at least one element of the feature data is derived from one or more of a time at which a corresponding sample value was collected or a location of the at least one playback device when the correspond sample value was collected.

3. The method of any preceding claim, wherein the current feature data is representative of one or more of a current time, a current user interaction with the one or more playback devices, or current location of the one or more playback devices.

4. The method of any preceding claim, further comprising: collecting, over time, the sample values of the one or more settings at the media playback system; and extracting the associated feature data from the sample values.

5. The method of claim 4, wherein collecting the sample values comprises collecting, by one or more devices of the media playback system, the sample values over time.

6. The method of any preceding claim, further comprising detecting input data representative of the current feature data.

7. The method of any preceding claim, wherein causing the at least one playback device to perform the at least one playback device operation includes adjusting the one or more settings to the recommended value.

8. The method of any preceding claim, wherein causing the at least one playback device to execute the at least one playback device operation includes providing, via the at least one playback device, one or more of an audible suggestion or a visual suggestion to adjust the setting to the recommended value.

9. The method of any preceding claim, wherein the one or more settings includes a volume level of at least one of the one or more playback devices.

10. The method of claim 3 alone or in combination with any other claim, wherein the current user interaction with the at least one playback devices includes one or more of a user command to adjust the volume level of at least one playback device or a user command to begin playback of audio content.

11. The method of any preceding claim, wherein the current feature data includes a group status of at least one playback device.

12. The method of any preceding claim, wherein the one or more settings includes a grouping configuration of one or more playback devices of the media playback system.

13. The method of claim 12, wherein the current feature data comprises a location of at least one playback device, and wherein executing the playback device operation comprises configuring a grouping configuration of a plurality of playback devices.

14. The method of claim 13, wherein configuring the grouping configuration comprises at least one of: automatically grouping the plurality of playback devices or automatically ungrouping the at least one playback device from a remainder of the plurality of playback devices.

15. The method of any preceding claim, wherein the parameterized machine learning model generates a confidence metric corresponding to the recommended value.

16. The method of claim 15, wherein causing the at least one playback device to perform the at least one playback device operation based on the recommendation and the confidence metric comprises one or more of: adjusting the one or more settings to the recommended value based on the confidence metric exceeding a threshold value; or causing the at least one playback device to output a recommendation to adjust the setting to the recommended value based on the confidence metric being at or below the threshold value.

17. The method of any preceding claim, wherein the parameterized machine learning model is a Gaussian Process model.

18. The method of any preceding claim, wherein training the parameterized machine learning model includes selecting, based on the sample values, a chosen parameterized machine learning model from among a plurality of parameterized machine learning models.

19. A network device comprising: at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor to control the network device to implement the method of any one of claims 1-18.

20. A media playback system comprising: at least one playback device; and one more devices configured for performing the method of any one of claims 1 to 18.