CN113160815A - Intelligent control method, device and equipment for voice awakening and storage medium - Google Patents
Intelligent control method, device and equipment for voice awakening and storage medium Download PDFInfo
- Publication number
- CN113160815A CN113160815A CN202110407187.7A CN202110407187A CN113160815A CN 113160815 A CN113160815 A CN 113160815A CN 202110407187 A CN202110407187 A CN 202110407187A CN 113160815 A CN113160815 A CN 113160815A
- Authority
- CN
- China
- Prior art keywords
- awakening
- current
- threshold
- smart device
- confidence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/4401—Bootstrapping
- G06F9/4418—Suspend and resume; Hibernate and awake
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Computer Security & Cryptography (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Electric Clocks (AREA)
Abstract
The application discloses intelligent control method, device, equipment and storage medium that pronunciation awaken up, through when the smart machine is awaken up, judge whether awaken up at present for the mistake awakens up, if for the mistake awakens up, then carry out the dormancy operation to the smart machine, and adjust the numerical value of the confidence coefficient threshold value of smart machine to first threshold value, thereby through improving the intelligent machine and awakening up the degree of difficulty under noisy environment, reduce the mistake awakening rate of smart machine, simultaneously, through carrying out the dormancy operation to the smart machine, reduce the energy consumption of smart machine.
Description
Technical Field
The present application relates to the field of voice processing technologies, and in particular, to a voice wake-up intelligent control method, apparatus, device, and storage medium.
Background
The voice awakening is a form of voice recognition technology, and the intelligent device can be awakened and operated through voice without directly contacting hardware equipment, so that the operation of a user is facilitated, and the intelligent device is not required to be in a working state in real time by adopting a voice awakening mechanism, so that the energy consumption can be saved.
However, in practical applications, various factors affect the accuracy of voice wake-up of the smart device, for example, when the smart device is in a noisy environment, the smart device is easily mistakenly woken up by a disturbing voice in the environment, which results in a high false wake-up rate of the smart device.
Disclosure of Invention
In view of the foregoing, the present invention provides a method, an apparatus, a device and a storage medium for intelligent voice wake-up control to improve the foregoing problems.
In a first aspect, an embodiment of the present application provides an intelligent control method for voice wakeup, where the method includes: when the intelligent device is awakened, whether the current awakening is mistaken awakening is judged. And when the confidence coefficient of the current awakening is larger than the confidence coefficient threshold value of the intelligent device, the intelligent device is awakened. And if the intelligent device is mistakenly awakened, executing the sleep operation on the intelligent device, and adjusting the value of the confidence coefficient threshold of the intelligent device to a first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
In a second aspect, an embodiment of the present application further provides an intelligent control device awakened by voice, where the device includes: a false wake-up confirmation module and a confidence threshold adjustment module. The false wake-up confirming module is used for judging whether the current wake-up is the false wake-up or not when the intelligent device is woken up. And when the confidence coefficient of the current awakening is larger than the confidence coefficient threshold value of the intelligent device, the intelligent device is awakened. The confidence threshold adjusting module is used for executing the sleep operation on the intelligent equipment if the current awakening is the false awakening, and adjusting the value of the confidence threshold of the intelligent equipment to a first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
In a third aspect, an embodiment of the present application further provides an electronic device, which includes one or more processors, a memory, and one or more application programs. Wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors. One or more programs configured to execute to implement the method of the first aspect as described above.
In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, where program codes are stored in the computer-readable storage medium. The program code may be invoked by a processor to perform a method as described in the first aspect above.
According to the technical scheme provided by the invention, when the intelligent equipment is awakened, whether the current awakening is mistaken awakening is judged, if the current awakening is mistaken awakening, the sleeping operation on the intelligent equipment is executed, and the numerical value of the confidence coefficient threshold value of the intelligent equipment is adjusted to the first threshold value, so that the mistaken awakening rate of the intelligent equipment is reduced by improving the awakening difficulty of the intelligent equipment in a noisy environment, and meanwhile, the energy consumption of the intelligent equipment is reduced by executing the sleeping operation on the intelligent equipment.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the description of the embodiments will be briefly introduced below, and it is apparent that the drawings in the following description are only some embodiments, not all embodiments, of the present application. All other embodiments and drawings obtained by a person skilled in the art based on the embodiments of the present application without any inventive step are within the scope of the present invention.
Fig. 1 is a schematic flowchart illustrating an intelligent voice wake-up control method according to an embodiment of the present application;
fig. 2 is a schematic flowchart illustrating an intelligent voice wake-up control method according to another embodiment of the present application;
fig. 3 is a schematic flowchart illustrating an intelligent voice wake-up control method according to another embodiment of the present application;
fig. 4 is a schematic flowchart illustrating an intelligent voice wake-up control method according to another embodiment of the present application;
fig. 5 is a schematic flowchart illustrating an intelligent voice wake-up control method according to another embodiment of the present application;
fig. 6 shows a block diagram of a voice-awakened intelligent control device according to an embodiment of the present application;
fig. 7 is a block diagram illustrating an electronic device according to an embodiment of the present application;
fig. 8 shows a block diagram of a computer-readable storage medium according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
The voice awakening technology has wide application field, can be applied to intelligent equipment such as robots, mobile phones, wearable equipment, smart homes and vehicles, and is used as a start or entrance for interaction between people and the intelligent equipment, so that infinite possibility is brought to intelligent life.
Generally, the smart device acquires a voice signal, analyzes a confidence level of the voice signal, and executes a wake-up operation on the smart device if the confidence level of the voice signal is greater than a preset confidence level threshold. However, in practical applications, when the smart device is in a noisy environment such as a multi-person conversation, the smart device is likely to be mistakenly awakened by the interfering voice in the environment, which results in a higher false awakening rate of the smart device.
In order to solve the above problem, the inventor of the present application provides a voice wake-up intelligent control method, apparatus, device and storage medium, which are provided by the present application, and determines whether the current wake-up is a false wake-up when the smart device is woken up, and if the current wake-up is a false wake-up, the sleep operation on the smart device is executed, and the value of the confidence threshold of the smart device is adjusted to the first threshold, so that the wake-up difficulty of the smart device in a noisy environment is increased, and the false wake-up rate of the smart device is reduced, and meanwhile, the energy consumption of the smart device is reduced by executing the sleep operation on the smart device.
Embodiments of the present application will be described in detail below with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present application provides an intelligent control method for voice wakeup, which can be applied to an electronic device with a voice wakeup function, and the present embodiment describes a flow of steps of an electronic device side, where the method may include steps S110 to S120.
Step S110, when the intelligent device is awakened, judging whether the current awakening is mistaken awakening. And when the confidence coefficient of the current awakening is larger than the confidence coefficient threshold value of the intelligent device, the intelligent device is awakened.
As an embodiment, the smart device may have a plurality of operating modes, such as a sleep mode, an awake mode, a power-off mode, and the like. It is to be appreciated that the smart device can be in a sleep mode when the smart device is not awake. In the sleep mode, the smart device may listen for ambient voice signals.
In some embodiments, the smart device picks up voice signals from the surrounding environment through its own pickup device (e.g., a microphone).
In other embodiments, the smart device may also obtain voice signals from other devices. For example, as one mode, after a sound pickup device of the peripheral device picks up a voice signal, the voice signal is sent to the smart device, and the smart device acquires the voice signal.
When the smart device receives a voice signal, a confidence level of the voice signal, i.e., a confidence level of the current wake-up, may be determined. If the confidence level of the received voice signal is greater than the confidence level threshold of the smart device, the smart device may be awakened based on the voice signal, that is, the smart device is switched from the sleep mode to the awake mode. The confidence threshold may be an initial confidence threshold of the smart device, or may be a confidence threshold obtained by adjusting a value of the initial confidence threshold of the smart device, that is, a confidence threshold of the smart device at the current time. In practical application, the initial confidence threshold of the intelligent device can be determined according to the user requirement and the practical application environment.
As an embodiment, the confidence level may use a wake-up model (or a pre-trained model) to analyze the input voice signal and determine the confidence level of the voice signal. Factors affecting the confidence may vary depending on the model selected. The factors affecting the confidence include, but are not limited to, acoustic features of the voice signal, such as included keywords (whether including a wakeup word), volume level (whether the volume reaches a preset value), and the like. In practical application, the confidence level determination mode can be selected according to the requirements of the application environment.
In some embodiments, the wake-up model may be a wake-up word detection model, and the wake-up word detection model determines whether a preset wake-up word exists in the voice signal. Alternatively, the wake word detection model may be pre-trained over a large number of training speech signals. And calculating the initial confidence coefficient of the voice signal through the awakening word detection model. In some embodiments, the confidence level of the voice signal refers to the similarity of the acoustic features of the wake-up word and the preset wake-up word. Alternatively, the wake model may be constructed by using a Convolutional Neural Network (CNN) algorithm, a Deep Neural Network (DNN) algorithm, or a Convolutional Recurrent Neural Network (CRNN) algorithm. It will be appreciated that the invention is not so limited and other wake-up models may be employed to obtain confidence in the speech signal.
It can be understood that the user wakes up the smart device to make the smart device enter a wake-up mode, so that the smart device can interact with the smart device to make the smart device complete the related instruction sent by the user. For example, the user wakes up the smart device for playing music, controlling the operating status of other devices, querying weather conditions, etc. And when the intelligent device is awakened due to noise, the intelligent device does not receive a voice command any more although the intelligent device enters an awakening mode.
Although the smart device wakes up after receiving the voice signal with the confidence level greater than the confidence level threshold, there is a certain false wake-up rate due to the method used by the smart device to calculate the confidence level. Thus, not every voice signal that causes a wake-up originates from a user with an interactive intent.
In some cases, noise in the environment may cause a wake-up of the smart device. The noise in the environment may be noise when other devices are operating, such as music playing devices, televisions, washing machines, etc. The noise in the environment may also be irregular noise, such as: noise generated during decoration and noise generated by vehicles passing by the road. In other cases, the smart device may wake up in a noisy environment, for example, in a multi-person conversation environment, a lot of noise may be generated, which may cause the smart device to wake up.
In the embodiment of the application, whether the current awakening is mistaken awakening can be judged by judging whether further interactive voice is received after the intelligent device is awakened.
And step S120, if the intelligent device is mistakenly awakened, executing the sleep operation on the intelligent device, and adjusting the value of the confidence coefficient threshold of the intelligent device to a first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
In some noisy environments, a significant amount of noise may be present in the environment for a period of time. For example, in a multi-person conversation scenario, there is a lot of noise in the user conversation in a short time environment, and the noise frequently causes false wake-up of the smart device in a short time. In the prior art, effective measures are not taken for the situation, so that the intelligent equipment is frequently and mistakenly awakened by noise, and the use experience of a user is seriously influenced.
In an embodiment of the application, when the current wake-up is recognized as the false wake-up, the sleep operation of the intelligent device is executed. Meanwhile, in order to enable the intelligent device not to be frequently awakened by the noise in a noisy environment, the value of the confidence coefficient threshold value of the intelligent device can be increased, so that the difficulty of awakening the intelligent device is increased, and the probability of mistaken awakening of the intelligent device caused by the noise in the environment is reduced.
In some embodiments, in order to increase the difficulty of the smart device being awakened in a noisy environment, the value of the confidence threshold of the smart device may be adjusted to a first threshold, and the first threshold is greater than the confidence of the current awakening. It can be understood that the confidence of the current wake-up is greater than the value of the original confidence threshold of the smart device, and therefore, the first threshold is also greater than the value of the original confidence threshold of the smart device, so that the difficulty of waking up the smart device is increased, and the false wake-up of the smart device is reduced.
According to the intelligent control method for voice awakening, when the intelligent device is awakened, whether the current awakening is mistaken awakening is judged, if the current awakening is mistaken awakening, the sleep operation on the intelligent device is executed, and the value of the confidence coefficient threshold value of the intelligent device is adjusted to the first threshold value, so that the mistaken awakening rate of the intelligent device is reduced by improving the awakening difficulty of the intelligent device in a noisy environment, and meanwhile, the energy consumption of the intelligent device is reduced by executing the sleep operation on the intelligent device.
Referring to fig. 2, another embodiment of the present application provides an intelligent control method for voice wakeup, which can be applied to an electronic device, and this embodiment describes a process of steps at the electronic device side, where the method can include steps S210 to S230.
And step S210, when the intelligent device is awakened, acquiring the current environment state information. Wherein the current environmental status information includes: and at least one of the working state of the audio playing device and the detected audio of the current environment.
It is understood that noise can cause false wake-up of the smart device. However, in some scenes, only a small amount of noise exists, so that the intelligent device is not frequently awakened by mistake, and the influence on the user is small. In some noisy scenes, a large amount of noise exists in the environment, which frequently causes false awakening of the intelligent device, and has a large influence on the user.
In order to more accurately and effectively suppress the false wake-up condition of the smart device in the noisy environment with a large influence, in the embodiment of the present application, when the smart device is woken up, it may be determined whether the current environment belongs to the noisy environment that needs to be suppressed.
In some embodiments, in order to determine whether the current environment belongs to a noisy environment that needs to be suppressed, the current environment state information may be obtained first.
In some embodiments, the current environmental status information includes an operational status of the audio playback device. It can be understood that the audio playing device generates a large amount of interference noise when playing audio. If the audio playing device in the surrounding environment of the intelligent device is in an audio playing state, a large amount of interference noise exists in the surrounding environment of the intelligent device, and the probability that the intelligent device is mistakenly awakened is high.
Optionally, the smart device and the audio playing device may be in communication connection via a network. In some ways, the smart device may obtain the operating state of the audio playback device by sending a query instruction to the audio playback device. In some manners, the audio playing device may also periodically send the operating state to the smart device, so that the smart device obtains the operating state of the audio playing device.
Alternatively, the Network described above is typically the internet, but may be any Network including, but not limited to, a Local Area Network (LAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a mobile, wireline or wireless Network, a private Network, or any combination of virtual private networks. In some embodiments, data exchanged over a network is represented using techniques and/or formats including Hypertext Mark-up Language (HTML), Extensible Markup Language (XML), and the like. All or some of the links may also be encrypted using conventional encryption techniques such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec). In other embodiments, custom and/or dedicated data communication techniques may also be used in place of, or in addition to, the data communication techniques described above.
In some implementations, the current environment state information includes detected audio of the current environment. It can be understood that the intelligent device can periodically collect the detection audio of the current environment, and further analyze the detection audio to determine the state of the current environment.
It is to be understood that the present application is not limited thereto, and other methods that can be used to determine the current environmental state may also be applied to the embodiments of the present application.
Step S220, if the current environment state information meets the preset condition, determining whether the current wake-up is a false wake-up.
In the embodiments of the present application, the preset condition refers to a condition that the current environment conforms to a noisy environment. It is understood that, when the current environmental status information is different, the preset condition needs to be set corresponding to the current environmental status information, which will be described in detail below.
In some embodiments, when the current environment state information is an operating state of the audio device, the preset condition is that the operating state of the audio device is an audio playing state. It will be appreciated that a significant amount of noise is generated when the audio device is in an audio playback state, such as when a music player plays music, a television is playing a program, etc.
In some embodiments, when the current environmental status information is the detection audio, whether the current environmental status information meets the preset condition may be determined by further analyzing the detection audio.
Alternatively, the detection audio may be a plurality of sample time sampled audio. By separately detecting parameters of each audio, for example: volume, signal to noise ratio, etc. As one way, if a parameter of a plurality of audios with a preset ratio in the audios is greater than a preset value, it is detected that the audios meet a preset condition. For example: if 4 sampling audios are collected, the audio is a first audio, a second audio, a third audio and a fourth audio, if the volume of the first audio is 60, the volume of the second audio is 70, the volume of the third audio is 70 and the audio of the fourth audio is 80, if it is assumed that the volume of the audio exceeding 60% exceeds 65%, the current environment state information meets the preset condition. It can be concluded that 75% of the four sampled audios have an audio volume exceeding 65, i.e., the current environmental status information meets the preset condition. It is to be understood that the present application is not limited thereto, and the number of sampled audios, the update time, the parameters of the audios, the preset conditions, and the like may be set according to the requirements of the actual application environment, and the present application is not limited thereto.
Optionally, whether a sound source in a fixed direction exists in the current environment may be determined by detecting the audio, and if so, the detected audio meets a preset condition. In some embodiments, the orientation of the sound in the detected audio may be obtained in a plurality of ways, and it is statistically confirmed whether there is a sound source in a fixed direction. For example: if 4 sampling audios are collected, the audio source direction of the fifth audio is in the first angle range, the audio source direction of the sixth audio is in the first angle range, the audio source direction of the seventh audio is in the first angle range, and the audio source direction of the eighth audio is in the first angle range. If it is assumed that more than 70% of the sound source directions of the audio are in the same angle range, the current environmental state information meets the preset condition. It can be found that 100% of the four sampled audios have a sound source bearing in the first angle range, and an interference sound source with a fixed bearing exists in the current environment, that is, the current environment state information meets the preset condition. It can be understood that the present application is not limited thereto, the number of the sampled audios, the update time, the preset conditions, and the like may be set according to the requirements of the actual application environment, and when determining whether there is an interfering sound source, the determination may also be performed in combination with the volume of the sound source, that is, except for a sound source with a fixed orientation, the volume of the sound source exceeds a preset proportion to meet the preset conditions, and the present application does not limit this.
In some embodiments, the obtained detection audio may be further input into an environmental state analysis model, and whether the detection audio meets a preset condition is determined by an environmental state analysis module.
Alternatively, the environment state analysis model may be obtained by training through a neural network model and through a large number of training data presets. In some embodiments, the training data may be a large amount of noisy audio for a multi-person conversation, so that the environmental state analysis module may analyze whether the acquired detection audio belongs to the detection audio in the multi-person noisy environment, and if so, the detection audio meets the preset condition.
In the embodiment of the application, when the current environment state information accords with the preset condition, the current environment is judged to be the noisy environment, the probability that the intelligent device is frequently awakened by mistake is higher, and whether the current awakening is mistaken awakening is further judged, so that the mistaken awakening suppression is pertinently carried out on the noisy environment.
Step S230, if the wake-up is false, performing a sleep operation on the smart device, and adjusting a confidence threshold of the smart device to a first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
For detailed description of step S230, please refer to step S120, which is not described herein again.
The intelligent control method for voice wake-up provided by the embodiment is improved on the basis of the previous embodiment, and the main improvement is as follows: when the intelligent device is awakened, the current environment state information is firstly acquired, and whether the current awakening is mistaken awakening or not is judged according to the current environment meeting the preset condition, so that the awakening difficulty of the intelligent device in the noisy environment is accurately improved, and the mistaken awakening rate of the intelligent device is effectively inhibited.
Referring to fig. 3, another embodiment of the present application provides an intelligent control method for voice wakeup, which can be applied to an electronic device, and this embodiment describes a process of steps at the electronic device side, where the method can include steps S310 to S320.
Step S310, when the intelligent device is awakened, if the voice signal of the user is not received within the preset detection time, determining that the current awakening is false awakening.
It can be understood that after the user wakes up the intelligent device normally, the user further issues a voice command to control the intelligent device to complete a corresponding command. If the smart device is awakened due to noise, the smart device does not necessarily receive further voice instructions.
In the embodiment of the application, when the smart device is awakened, in order to further determine whether the current awakening is mistaken awakening, whether the smart device further receives the voice signal of the user may be determined.
In the embodiment of the application, in the preset detection time when the intelligent device is awakened, if the voice signal of the user is not received, it is determined that no further interactive instruction is obtained after the intelligent device is awakened, and it can be determined that the current awakening is mistaken awakening caused by noise. The preset detection time can be adjusted according to the actual application scene.
Step S320, if the wake-up is false, executing a sleep operation on the smart device, and adjusting the confidence threshold of the smart device to the first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
For a detailed description of step S320, please refer to step S120, which is not described herein again.
The intelligent control method for voice wake-up provided by the embodiment is improved on the basis of the previous embodiment, and the main improvement is as follows: when the intelligent device is awakened, whether the current awakening is mistaken awakening is determined by whether the voice signal of the user is received within preset detection time, and the mistaken awakening rate of the intelligent device is reduced by improving the awakening difficulty of the intelligent device in a noisy environment.
Referring to fig. 4, another embodiment of the present application provides an intelligent control method for voice wakeup, which can be applied to an electronic device, and this embodiment describes a flow of steps at the electronic device side, where the method can include steps S410 to S430.
It will be appreciated that if the smart device is awakened due to noise, the smart device will not receive further user voice commands, except as mentioned in step S310 of the above embodiment. In other embodiments, after the smart device is awakened due to noise, the smart device may further receive the voice signal, but the received voice signal is irrelevant to the interaction, and the current awakening may be determined as false awakening. As will be specifically explained below.
Step S410, when the intelligent device is awakened, if the voice signal of the user is received within the preset detection time, performing intention recognition on the voice signal.
Step S420, judging whether the current awakening is mistaken awakening according to the intention identification result.
In the embodiment of the application, when the intelligent device is awakened, if the voice signal of the user is received within the preset detection time, in order to determine whether the current awakening is mistaken awakening, intention recognition can be performed on the received voice signal.
Intent recognition is the determination of the user's intent, i.e., what the user wants to do. Knowing what the user wants to do can correspond to executing the user's instructions. In some embodiments, after receiving the voice command, the smart device may identify the voice signal and perform intent recognition to obtain the intent of the user corresponding to the voice signal. The intention of the user corresponding to the voice signal is used for reflecting an operation that the user desires to be performed by the smart device, for example, the intention may include "turn on a television", "play music", "inquire weather", and the like, which is not limited herein. For example, when the voice signal is "playing sunny days", it may be determined that the voice signal corresponds to an intention of "playing music". When the speech signal is "today's weather", it may be determined that the speech signal corresponds to the intention of "music playing", and the like, which is not limited herein.
In some embodiments, the voice signal may be input to an intention classification model for intention classification, and an intention corresponding to the voice signal may be obtained. The entity in the voice signal can be extracted and input to the intention classification model for intention classification, and the intention corresponding to the voice signal is obtained. Or extracting an entity in the voice signal, acquiring entity content of the entity based on a knowledge graph, inputting the entity content into an intention classification model for intention classification, acquiring an intention corresponding to the voice signal, and the like. The intention classification model can be obtained by training classification models such as an SVM (Support Vector Machine) model, a neural network model or a random forest model, user intention recognition is achieved through Machine learning, and accuracy of user intention recognition is effectively improved.
In the embodiment of the application, if it is recognized that the intention included in the voice signal of the user is the target intention, the smart device may execute a corresponding operation. Target intent refers to the type of intent that the smart device can handle. When the voice signal contains a target intention, namely the user sends out the voice signal with a definite intention, the voice signal indicates that the user has an intention of interacting with the intelligent device, namely the current wake-up is the wake-up actively triggered by the user for interacting with the intelligent device. When the voice signal does not contain the target intention, the current awakening is not the environment actively triggered by the user, but the false awakening caused by the noise in the environment.
In some implementations, the target intent of the smart device does not include a chat intent. I.e. the user's speech signal contains an intention that has a well-defined control requirement and that the smart device can support. Such as playing music, querying weather, turning on air conditioning, etc.
In some implementations, the target intent of the smart device can also include a chat intent. If the intention contained by the speech signal cannot be recognized, for example, the user's speech signal is "eaten", "haha", "too heavy", etc. At this time, the smart device may open a chatting mode to perform chatting with the user.
Optionally, if the smart device does not receive the feedback voice of the user within a preset time after the smart device opens the chatting mode, the user does not have an intention of chatting with the smart device, and it may be confirmed that the voice signal of the user does not include the target intention.
Optionally, if the smart device receives feedback voice of the user within a preset time after the smart device starts the chatting mode, the intention included in the voice signal may be classified as the chatting intention. For example, the voice received from the user is "too heavy", the smart device turns on a chatting mode "what you have taken, and the feedback voice of the user is" i have taken too much to eat ".
Further, in order to reduce interference of noise on the chatting intention recognition, after the intelligent device receives the feedback voice, the correlation degree between the feedback voice and the chatting content of the intelligent device can be further judged, and if the correlation degree is greater than a preset value, the voice signal of the user is judged to contain the target intention. Optionally, the calculation of the correlation degree of the chatting content may be performed by inputting the feedback voice and the chatting content of the intelligent device into a pre-trained neural network model for calculation, and the preset value may be set according to the actual application requirement, which is not limited in this application.
In some embodiments, if only one voice signal is received within the preset detection time and the intention recognition result of the voice signal is that the target intention is not recognized, the current wake-up is determined to be a false wake-up after the preset detection time. It can be understood that, if the intention recognition result of the received voice signal is that the target intention is recognized, it is determined that the current wake-up is not a false wake-up, and then the corresponding operation is executed according to the intention included in the voice signal.
In some embodiments, if a plurality of voice signals are received within a preset detection time and the target intention is not recognized in all the intention recognition results of the plurality of voice signals, it is determined that the current wake-up is a false wake-up.
Under some noisy environments, the intelligent device can receive a large amount of noises after being awakened, and in order to further reduce the influence of the noises on the intelligent device, the intelligent device can be set to be awakened into a false awakening mode after receiving preset number of voice signals which do not contain target intentions. For example, if the preset detection time is 10 seconds and the preset number is 3, if the intelligent device receives the first voice in the 1 st second, the intention recognition result of the first voice is that the target intention is not recognized; receiving a second voice in the 3 rd second, wherein the intention recognition result of the second voice is that the target intention is not recognized; and receiving the third voice in the 5 th second, wherein the intention recognition result of the third voice is that the target intention is not recognized, and judging that the current awakening is mistaken awakening in the 5 th second.
And step S430, if the intelligent device is mistakenly awakened, executing the sleep operation on the intelligent device, and adjusting the value of the confidence coefficient threshold of the intelligent device to a first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
For detailed description of step S430, please refer to step S120, which is not described herein again.
The intelligent control method for voice wake-up provided by the embodiment is improved on the basis of the previous embodiment, and the main improvement is as follows: when the intelligent device is awakened, if the voice signal of the user is received within the preset detection time, intention identification is carried out on the voice signal, whether current awakening is mistaken awakening or not is judged according to an intention identification result, and the mistaken awakening rate of the intelligent device is reduced by improving the awakening difficulty of the intelligent device in a noisy environment.
Referring to fig. 5, another embodiment of the present application provides an intelligent control method for voice wakeup, which is applicable to an electronic device, and the present embodiment describes a process of steps at the electronic device side, where the method may include steps S510 to S550.
Step S510, when the smart device is awakened, determining whether the current awakening is false awakening. And when the confidence coefficient of the current awakening is larger than the confidence coefficient threshold value of the intelligent device, the intelligent device is awakened.
Step S520, if the false wake-up is detected, performing a sleep operation on the smart device, and adjusting the confidence threshold of the smart device to a first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
For the detailed description of steps S510 to S520, refer to steps S110 to S120, which are not described herein again.
Step S530, if the recovery time is preset, the voice signal of the user is received, and the confidence coefficient of the voice signal is determined.
After step S520 is executed, the smart device is in the sleep mode, and the value of the confidence threshold of the smart device is increased, that is, the difficulty of waking up the smart device is increased, so that the false waking up of the smart device by a part of noise can be effectively isolated. Alternatively, the preset resume time may be calculated from when the hibernation operation for the smart device is performed. Optionally, the preset recovery time may also be calculated from when the confidence threshold of the smart device is adjusted. The length of the preset recovery time can be adjusted according to the actual application scene.
In the preset recovery time, if the voice signal of the user is received, the confidence level of the voice signal is determined to determine whether the received voice signal can wake up the intelligent device, and the calculation of the confidence level of the voice signal may refer to the description of the foregoing embodiment, which is not described herein again.
And step S540, if the confidence of the voice signal is greater than the current confidence threshold, performing awakening operation on the intelligent device based on the voice signal.
It can be understood that, after the value of the confidence threshold of the intelligent device is increased, the waking difficulty of the intelligent device is increased, and if the confidence of the received voice signal is greater than the current confidence threshold, the voice signal is a normal wake of the user, that is, the user has a need to control the intelligent device, and then the waking operation on the intelligent device is executed based on the voice signal.
In some embodiments, if the confidence level of the voice signal is smaller than the current confidence level threshold, the voice signal is noise, the smart device is still in the sleep mode, and the smart device continues to monitor whether other voice signals are input.
And step S550, adjusting the value of the confidence coefficient threshold of the intelligent device from a first threshold to a second threshold, wherein the second threshold is smaller than the first threshold.
In some embodiments, the value of the confidence threshold of the smart device is adjusted from a first threshold to a second threshold when the smart device is awakened within a preset recovery time. The second threshold is smaller than the first threshold, namely, after the intelligent device is awakened within the preset recovery time, the difficulty of awakening the intelligent device is reduced. Preferably, the first threshold is a default value of the confidence threshold in a normal working state of the intelligent device.
In some embodiments, when the smart device is awakened within the preset recovery time, it may be further determined whether the awakening is false awakening. And adjusting the value of the confidence coefficient threshold of the intelligent equipment according to the awakening result.
Optionally, when the smart device is awoken within the preset recovery time, the value of the confidence threshold of the smart device may be adjusted to be higher, for example, the value of the confidence threshold of the smart device is adjusted to be a third threshold, where the third threshold is greater than the first threshold. Therefore, the difficulty of awakening the intelligent equipment is continuously increased, and the false awakening rate of the intelligent equipment is reduced.
It can be understood that after the value of the confidence threshold of the smart device is increased, the waking difficulty of the smart device is increased, which can prevent the smart device from being mistakenly woken by part of noise, but also increases the difficulty of the user in waking up the smart device. However, the smart device is usually in a noisy environment for a short time, for example, a multi-person conversation scene is often found in the morning or evening. If the intelligent device is always in a state of high awakening difficulty, normal use of the user can be influenced. Therefore, in order to be closer to the actual use requirement, as an implementation manner, if the smart device is not awakened within the preset recovery time, that is, the voice signal of the user is not received within the preset recovery time, the value of the confidence threshold of the smart device is adjusted after the preset recovery time, so that the value of the confidence threshold of the smart device is reduced, and the difficulty of awakening the smart device is reduced. Preferably, the confidence threshold of the smart device may be adjusted from a first threshold to a second threshold, or to an initial confidence.
The intelligent control method for voice wake-up provided by the embodiment is improved on the basis of the previous embodiment, and the main improvement is as follows: after the awakening difficulty of the intelligent device in the noisy environment is improved, if the awakening difficulty is within the preset recovery time, the voice signal of the user is received, and the confidence coefficient of the voice signal is larger than the current confidence coefficient threshold value, the numerical value of the confidence coefficient threshold value of the intelligent device is adjusted from the first threshold value to the second threshold value, so that when the intelligent device is needed to be used by the user, the confidence coefficient threshold value of the intelligent device is rapidly identified and recovered, and the use experience of the user is improved.
Referring to fig. 6, which illustrates a voice-awakened intelligent control device 600 according to an embodiment of the present invention, the voice-awakened intelligent control device includes: a false wake confirmation module 610 and a confidence threshold adjustment module 620.
Specifically, the false wake up confirmation module 610 is configured to determine whether the current wake up is a false wake up when the smart device is woken up. And when the confidence coefficient of the current awakening is larger than the confidence coefficient threshold value of the intelligent device, the intelligent device is awakened.
The confidence threshold adjustment module 620 is configured to, if the smart device is awoken by mistake, perform a sleep operation on the smart device, and adjust a value of the confidence threshold of the smart device to a first threshold. Wherein the first threshold is greater than the confidence of the current wake-up.
Furthermore, the intelligent control device for voice awakening further comprises a current environment state information acquisition module and a false awakening judgment module.
The current environment state information acquisition module is used for acquiring current environment state information when the intelligent device is awakened. Wherein the current environmental status information includes: and at least one of the working state of the audio playing device and the detected audio of the current environment.
And the false wake-up judging module is used for judging that the current wake-up is the false wake-up if the current environment state information accords with the preset condition.
Furthermore, the intelligent control device for voice awakening further comprises a first voice signal determining module, a second voice signal determining module and a false awakening recognition module.
The first voice signal determining module is used for determining that the current awakening is false awakening if the voice signal of the user is not received within the preset detection time when the intelligent device is awakened.
The second voice signal determination module is used for performing intention recognition on the voice signal if the voice signal of the user is received within the preset detection time when the intelligent device is awakened.
And the false wake-up identification module is used for judging whether the current wake-up is false wake-up according to the intention identification result.
Further, the intelligent control device for semantic awakening further comprises: the device comprises a voice confirmation module, a confidence coefficient determination module and a confidence coefficient threshold value recovery module.
The voice confirmation module is used for receiving the voice signal of the user and determining the confidence coefficient of the voice signal if the voice signal is received within the preset recovery time.
The confidence level determining module is used for executing the awakening operation of the intelligent device based on the voice signal if the confidence level of the voice signal is greater than the current confidence level threshold value.
The confidence threshold restoration module is used for adjusting the value of the confidence threshold of the intelligent device from a first threshold to a second threshold, wherein the second threshold is smaller than the first threshold.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the device-like embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment. For any processing manner described in the method embodiment, all the processing manners may be implemented by corresponding processing modules in the apparatus embodiment, and details in the apparatus embodiment are not described again.
Referring to fig. 7, based on the above-mentioned voice-awakening intelligent control method, another electronic device 700 including a processor capable of executing the above-mentioned voice-awakening intelligent control method is further provided in the embodiment of the present application, where the electronic device 700 further includes one or more processors 710, a memory 720 and one or more application programs. The memory 720 stores programs that can execute the content of the foregoing embodiments, and the processor 710 can execute the programs stored in the memory 720. Wherein, the electronic device 700 may be an intelligent control panel, a smart phone, an intelligent wearable device, an intelligent voice navigation device, an intelligent robot, a tablet computer, a personal computer, or the like.
The Memory 720 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). The memory 720 may be used to store instructions, programs, code sets, or instruction sets. The memory may include a stored program area and a stored data area, where the stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as receiving voice, etc.), instructions for implementing the various method embodiments described below, and the like. The stored data area may also store data created by the terminal in use, such as current environmental state information, confidence thresholds, voice signals, etc.
Referring to fig. 8, a block diagram of a computer-readable storage medium 800 according to an embodiment of the present application is shown. The computer-readable storage medium 800 has stored therein a program code 810, said program code 810 being invokable by the processor for performing the method described in the above method embodiments.
The computer-readable storage medium 800 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium includes a non-volatile computer-readable storage medium. The computer readable storage medium has storage space for program code 810 for performing any of the method steps of the method described above. The program code 810 can be read from or written to one or more computer program products. The program code may be compressed, for example, in a suitable form.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. An intelligent control method for voice wake-up, the method comprising:
when the intelligent equipment is awakened, judging whether the current awakening is mistaken awakening or not, wherein when the confidence coefficient of the current awakening is larger than the confidence coefficient threshold value of the intelligent equipment, the intelligent equipment is awakened;
and if the intelligent device is mistakenly awakened, executing the sleep operation on the intelligent device, and adjusting the data of the confidence coefficient threshold of the intelligent device to a first threshold, wherein the first threshold is larger than the confidence coefficient of the current awakening.
2. The method of claim 1, wherein determining whether the current wake-up is a false wake-up when the smart device is woken up comprises:
when the intelligent equipment is awakened, acquiring current environment state information; wherein the current environmental state information includes: at least one of the working state of the audio playing device and the detected audio of the current environment;
and if the current environment state information meets the preset condition, judging whether the current awakening is mistaken awakening.
3. The method of claim 1, wherein determining whether the current wake-up is a false wake-up when the smart device is woken up comprises:
when the intelligent equipment is awakened, if the voice signal of the user is not received within the preset detection time, the current awakening is determined to be false awakening.
4. The method of claim 1, wherein determining whether the current wake-up is a false wake-up when the smart device is woken up comprises:
when the intelligent equipment is awakened, if a voice signal of a user is received within preset detection time, performing intention recognition on the voice signal;
and judging whether the current awakening is mistaken awakening according to the intention identification result.
5. The method of claim 4, wherein the determining whether the current wake-up is a false wake-up according to the intent recognition result comprises:
and if only one voice signal is received within the preset detection time or a plurality of voice signals are received within the preset detection time and the intention recognition result of one or more voice signals is that the target intention is not recognized, determining that the current awakening is mistaken awakening.
6. The method of claim 1, wherein after performing a sleep operation on the smart device and adjusting the confidence threshold value of the smart device to a first threshold value if the smart device is awoken, the method further comprises:
if the voice signal of the user is received within the preset recovery time, determining the confidence coefficient of the voice signal;
if the confidence of the voice signal is greater than the current confidence threshold, performing awakening operation on the intelligent equipment based on the voice signal;
adjusting a value of a confidence threshold of the smart device from the first threshold to a second threshold, wherein the second threshold is less than the first threshold.
7. The method of claim 1, wherein after performing a sleep operation on the smart device and adjusting the confidence threshold value of the smart device to a first threshold value if the smart device is awoken, the method further comprises:
and if the voice signal of the user is not received within the preset recovery time, recovering the numerical value of the confidence coefficient threshold value to the initial value.
8. An intelligent voice-awakening control device, comprising:
the false wake-up confirming module is used for judging whether the current wake-up is the false wake-up or not when the intelligent equipment is woken up; wherein the smart device is awakened when the confidence of the current awakening is greater than the confidence threshold of the smart device;
and the confidence threshold adjusting module is used for executing the sleep operation on the intelligent equipment if the intelligent equipment is awoken by mistake, and adjusting the value of the confidence threshold of the intelligent equipment to a first threshold, wherein the first threshold is larger than the current awakening confidence.
9. An electronic device, comprising:
one or more processors;
a memory;
one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the method of any of claims 1-7.
10. A computer-readable storage medium, having stored thereon program code that can be invoked by a processor to perform the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110407187.7A CN113160815B (en) | 2021-04-15 | 2021-04-15 | Intelligent control method, device, equipment and storage medium for voice wakeup |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110407187.7A CN113160815B (en) | 2021-04-15 | 2021-04-15 | Intelligent control method, device, equipment and storage medium for voice wakeup |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113160815A true CN113160815A (en) | 2021-07-23 |
| CN113160815B CN113160815B (en) | 2023-08-29 |
Family
ID=76867940
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110407187.7A Active CN113160815B (en) | 2021-04-15 | 2021-04-15 | Intelligent control method, device, equipment and storage medium for voice wakeup |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113160815B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113870856A (en) * | 2021-10-18 | 2021-12-31 | 深圳追一科技有限公司 | Public place digital people defense false wake-up method, related device and storage medium |
| CN114220423A (en) * | 2021-12-31 | 2022-03-22 | 思必驰科技股份有限公司 | Voice wake-up, method of customizing wake-up model, electronic device, and storage medium |
| CN115019798A (en) * | 2022-06-21 | 2022-09-06 | 珠海格力电器股份有限公司 | Control method, device, electronic device and storage medium for speech recognition device |
| CN115019835A (en) * | 2022-05-27 | 2022-09-06 | 江西省天轴通讯有限公司 | Intelligent management method and system for equipment, storage medium and equipment |
| CN115171699A (en) * | 2022-05-31 | 2022-10-11 | 青岛海尔科技有限公司 | Wake-up parameter adjusting method and device, storage medium and electronic device |
| CN115472161A (en) * | 2022-07-27 | 2022-12-13 | 北京声智科技有限公司 | Voice wake-up method, device, device and storage medium |
| CN116259314A (en) * | 2022-09-09 | 2023-06-13 | 青岛海尔空调器有限总公司 | Method and device for controlling voice-controlled equipment, controlling voice-controlled equipment |
| CN118522281A (en) * | 2024-05-31 | 2024-08-20 | 北京蜂巢世纪科技有限公司 | Trigger action recognition method and device, wearable device, and storage medium |
| CN119847620A (en) * | 2025-03-19 | 2025-04-18 | 南京汤峰机电有限公司 | Low-power-consumption standby intelligent wake-up control system and method for percussion drill |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110312285A1 (en) * | 2010-06-16 | 2011-12-22 | Essence Security International Ltd. | Adaptive thresholding in a wake-on-radio system |
| CN105721671A (en) * | 2016-01-18 | 2016-06-29 | 广东小天才科技有限公司 | Portable terminal dormancy method for mistaken awakening and portable terminal |
| CN109256134A (en) * | 2018-11-22 | 2019-01-22 | 深圳市同行者科技有限公司 | A kind of voice awakening method, storage medium and terminal |
| CN109346071A (en) * | 2018-09-26 | 2019-02-15 | 出门问问信息科技有限公司 | Wake up processing method, device and electronic equipment |
| CN109672775A (en) * | 2017-10-16 | 2019-04-23 | 腾讯科技（北京）有限公司 | Adjust the method, apparatus and terminal of wakeup sensitivity |
| CN111816178A (en) * | 2020-07-07 | 2020-10-23 | 云知声智能科技股份有限公司 | Voice equipment control method, device and equipment |
-
2021
- 2021-04-15 CN CN202110407187.7A patent/CN113160815B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110312285A1 (en) * | 2010-06-16 | 2011-12-22 | Essence Security International Ltd. | Adaptive thresholding in a wake-on-radio system |
| CN105721671A (en) * | 2016-01-18 | 2016-06-29 | 广东小天才科技有限公司 | Portable terminal dormancy method for mistaken awakening and portable terminal |
| CN109672775A (en) * | 2017-10-16 | 2019-04-23 | 腾讯科技（北京）有限公司 | Adjust the method, apparatus and terminal of wakeup sensitivity |
| CN109346071A (en) * | 2018-09-26 | 2019-02-15 | 出门问问信息科技有限公司 | Wake up processing method, device and electronic equipment |
| CN109256134A (en) * | 2018-11-22 | 2019-01-22 | 深圳市同行者科技有限公司 | A kind of voice awakening method, storage medium and terminal |
| CN111816178A (en) * | 2020-07-07 | 2020-10-23 | 云知声智能科技股份有限公司 | Voice equipment control method, device and equipment |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113870856A (en) * | 2021-10-18 | 2021-12-31 | 深圳追一科技有限公司 | Public place digital people defense false wake-up method, related device and storage medium |
| CN113870856B (en) * | 2021-10-18 | 2025-01-28 | 深圳追一科技有限公司 | A method for preventing false awakening of digital people in public places, related devices and storage medium |
| CN114220423A (en) * | 2021-12-31 | 2022-03-22 | 思必驰科技股份有限公司 | Voice wake-up, method of customizing wake-up model, electronic device, and storage medium |
| CN115019835A (en) * | 2022-05-27 | 2022-09-06 | 江西省天轴通讯有限公司 | Intelligent management method and system for equipment, storage medium and equipment |
| CN115171699A (en) * | 2022-05-31 | 2022-10-11 | 青岛海尔科技有限公司 | Wake-up parameter adjusting method and device, storage medium and electronic device |
| CN115019798A (en) * | 2022-06-21 | 2022-09-06 | 珠海格力电器股份有限公司 | Control method, device, electronic device and storage medium for speech recognition device |
| CN115472161A (en) * | 2022-07-27 | 2022-12-13 | 北京声智科技有限公司 | Voice wake-up method, device, device and storage medium |
| CN116259314A (en) * | 2022-09-09 | 2023-06-13 | 青岛海尔空调器有限总公司 | Method and device for controlling voice-controlled equipment, controlling voice-controlled equipment |
| WO2024051199A1 (en) * | 2022-09-09 | 2024-03-14 | 青岛海尔空调器有限总公司 | Method and apparatus for controlling voice control device, and device for controlling voice control |
| CN118522281A (en) * | 2024-05-31 | 2024-08-20 | 北京蜂巢世纪科技有限公司 | Trigger action recognition method and device, wearable device, and storage medium |
| CN119847620A (en) * | 2025-03-19 | 2025-04-18 | 南京汤峰机电有限公司 | Low-power-consumption standby intelligent wake-up control system and method for percussion drill |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113160815B (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113160815B (en) | Intelligent control method, device, equipment and storage medium for voice wakeup | |
| CN111968644B (en) | Intelligent device awakening method and device and electronic device | |
| CN108735209B (en) | Wake-up word binding method, intelligent device and storage medium | |
| CN108711430B (en) | Speech recognition method, intelligent device and storage medium | |
| CN113066488B (en) | Voice wakeup intelligent control method and device, electronic equipment and storage medium | |
| CN110364156A (en) | Voice interaction method, system, terminal and readable storage medium | |
| CN106157950A (en) | Speech control system and awakening method, Rouser and household electrical appliances, coprocessor | |
| CN110517670A (en) | Method and apparatus for improving wake-up performance | |
| CN111667818A (en) | Method and device for training awakening model | |
| CN105575395A (en) | Voice wake-up method and apparatus, terminal, and processing method thereof | |
| CN112233676B (en) | Intelligent device awakening method and device, electronic device and storage medium | |
| CN111312222A (en) | Awakening and voice recognition model training method and device | |
| CN112700782A (en) | Voice processing method and electronic equipment | |
| CN112669822B (en) | Audio processing method and device, electronic equipment and storage medium | |
| CN111161714A (en) | Voice information processing method, electronic equipment and storage medium | |
| CN113963695A (en) | Awakening method, awakening device, equipment and storage medium of intelligent equipment | |
| CN112669837B (en) | Awakening method and device of intelligent terminal and electronic equipment | |
| CN111722696B (en) | Voice data processing method and device for low-power-consumption equipment | |
| CN113241073B (en) | Intelligent voice control method, device, electronic equipment and storage medium | |
| CN112311635A (en) | Voice interrupt wake-up method, device and computer-readable storage medium | |
| CN113808584B (en) | Voice wakeup method and device, electronic equipment and storage medium | |
| CN112562742A (en) | Voice processing method and device | |
| CN111651135B (en) | Sound awakening method and device, storage medium and electrical equipment | |
| CN113889116A (en) | Voice information processing method and device, storage medium and electronic device | |
| CN112269322A (en) | Awakening method and device of intelligent device, electronic device and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |