WO2017071453A1 - Method and device for voice recognition - Google Patents

Abstract

A method and device for voice recognition. The method comprises: acquiring a state of an audio player of an electronic device (101); when the audio player is in a playback state, utilizing a first microphone of the electronic device to capture live sounds at the scene, and utilizing a second microphone of the electronic device to capture sounds being played back by the audio player (102), where the live sounds at the scene comprise a voice command of a user and the sounds played back by the audio player; recognizing the voice command of the user from the live sounds at the scene and the sound played back by the audio player (103); and operating the electronic device on the basis of the recognized voice command of the user (104). The technical solution provided in embodiments effectively solves the obstacle to voice recognition technology in a complex scenario, allows the user to use voice commands to effectively operate the electronic device while the audio player is in the playback state, thus greatly enhancing user experience.

Description

Method and device for speech recognition

The present application claims priority to Chinese Patent Application No. 201510716257.1, entitled "A Method and Apparatus for Speech Recognition", filed on October 28, 2015, the entire contents of in.

Technical field

The invention relates to the field of speech recognition technology, and in particular relates to a method and a device for speech recognition.

Background technique

At present, with the development of science and technology, people can control electronic devices through voice. In a quiet environment, the electronic device can effectively collect the user's voice, analyze and process the user's voice, in order to obtain a valid instruction, and act according to the instruction, thereby effectively responding to the user's voice.

However, when the electronic device is playing music, the user indicates the electronic device by voice. At this time, the electronic device acquires the user's voice and the played music at the same time, and the user's voice and the played music are mixed to correctly identify the user's voice. Adding a lot of difficulty.

Summary of the invention

The invention provides a method and a device for voice recognition, which effectively solves the obstacle of the voice recognition technology in a complex scene, so that the user can use the voice command to effectively operate the electronic device while the sound is in the playing state, which greatly improves the operation. User experience.

A first aspect of the embodiments of the present invention discloses a method for voice recognition, including:

Obtaining the state of the sound of the electronic device;

When the sound is in a playing state, the first microphone of the electronic device is used to collect the sound of the scene, and the second microphone of the electronic device is used to obtain the sound of the sound playing, wherein the sound of the scene includes the voice instruction of the user and The sound of the sound playing;

Identifying a user's voice command from the sound of the scene and the sound played by the sound;

The electronic device is operated according to the recognized voice instruction of the user.

In conjunction with the first aspect, in a first possible implementation manner of the first aspect, before the recognizing a voice instruction of the user from the sound of the scene and the sound of the sound playing, the method further includes:

The electronic device fills the sound of the scene into the first channel to obtain first audio data;

The electronic device fills the sound of the audio playback into the second channel to obtain second audio data.

With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the electronic device acquires a user from the sound of the scene and the sound played by the sound according to a preset method. Voice commands, including:

The electronic device acquires data of the first channel and data of the second channel; the electronic device uses a frequency conversion method from data of the first channel and data of the second channel Obtaining a valid audio data stream;

The electronic device performs noise cancellation on the valid audio data stream using an automatic gain control algorithm AGC to acquire a user's voice command.

In conjunction with the first possible implementation of the first aspect, in a third possible implementation of the first aspect, the method further includes:

The electronic device controls the delay of data between the first channel and the second channel to be less than a threshold by coordinating the first microphone and the second microphone.

In conjunction with the first aspect, in a fourth possible implementation of the first aspect, the method further includes:

When the sound of the electronic device is in a non-playing state, the electronic device acquires a user voice by using the first microphone and the second microphone;

The electronic device performs noise cancellation on the user voice by using an AGC algorithm to acquire a voice instruction of the user;

The electronic device operates in accordance with a voice instruction of the user.

With reference to the first aspect, in a fifth possible implementation manner of the first aspect, the first microphone is a primary microphone; the second microphone is a secondary microphone; the first channel is a left channel; The second channel is the right channel.

A second aspect of the embodiments of the present invention discloses a device for voice recognition, including:

a first acquiring unit configured to acquire a state of an audio of the electronic device;

The collecting unit is configured to collect the sound of the scene by using the first microphone of the electronic device when the sound is in the playing state, and acquire the sound of the sound playing by using the second microphone of the electronic device;

The sound of the scene includes a voice command of the user and a sound played by the sound;

a second acquiring unit, configured to acquire a voice instruction of the user from the sound of the scene and the sound played by the sound;

The operating unit is configured to operate the electronic device according to the recognized voice command of the user.

In conjunction with the second aspect, in a first possible implementation manner of the second aspect, the device further includes:

The processing unit is configured to: the electronic device fills the sound of the scene into the first channel to obtain the first audio data, and fills the sound of the audio to the second channel to obtain the second audio data.

With reference to the first possible implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the second acquiring unit is specifically configured to:

Obtaining data of the first channel and data of the second channel;

Acquiring a valid audio data stream from the data of the first channel and the data of the second channel by using a frequency conversion method;

The effective audio data stream is subjected to noise cancellation using an automatic gain control algorithm AGC to acquire a user's voice command.

In conjunction with the second aspect or the first possible implementation of the second aspect, in a third possible implementation of the second aspect, the device further includes:

And a control unit configured to: control, by the electronic device, the delay of data between the first channel and the second channel by using the first microphone and the second microphone to be less than a threshold.

In conjunction with the second aspect, in a fourth possible implementation of the second aspect,

The first obtaining unit is further configured to: when the sound is in a non-playing state, the electronic device acquires a user voice by using the first microphone and the second microphone;

The speech is noise cancelled using an AGC algorithm to obtain the user's voice instructions.

It can be seen that, in the solution of the embodiment of the present invention, when the electronic device audio is in the playing state, the electronic device may use the first microphone to collect the sound of the scene, and use the second microphone to obtain the sound of the audio playing, wherein the scene The sound of the user includes the voice command of the user and the voice played by the sound. According to the technical solution, the electronic device can remove the sound of the sound played in the live sound acquired by the first microphone according to the sound of the sound played by the second microphone, thereby The user voice command is obtained, so that the user can use the voice command to effectively operate the electronic device while the sound is in the playing state, thereby greatly improving the user experience.

DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings to be used in the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without paying any creative work.

1 is a schematic flowchart of a method for voice recognition according to an embodiment of the present invention;

FIG. 2 is a schematic flowchart diagram of another method for voice recognition according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a voice recognition apparatus according to an embodiment of the present invention; FIG.

4 is a schematic diagram of another voice recognition apparatus according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of another voice recognition apparatus according to an embodiment of the present invention; FIG.

FIG. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

detailed description

The invention provides a method and a device for voice recognition, which effectively solves the obstacle of voice recognition technology in a complex scene, so that the user can use the voice command to effectively operate the electronic device at the same time when the sound is in the playing state, which greatly improves the operation. User experience.

The technical solutions in the embodiments of the present invention will be clearly described in conjunction with the drawings in the embodiments of the present invention. Some embodiments of the invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of the present invention.

The terms used in the embodiments of the present invention are for the purpose of describing particular embodiments only and are not intended to limit the invention. The singular forms "a", "the" and "the" It will also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items. It is further understood that the term "comprising", used herein, is intended to mean the presence of the features, the whole, the steps, the operation, the elements and / , or the presence or addition of components and/or their groups

The method described in the embodiments of the present invention can be applied to various types of intelligent terminals having voice recognition functions, such as a tablet computer, a smart phone, an e-reader, a remote controller, a personal computer PC, a notebook computer, an in-vehicle device, a network television, and the like. A smart terminal having a voice recognition function such as a wearable device.

An embodiment of a method for voice recognition, a method for voice recognition, the method comprising: acquiring a state of an audio of an electronic device; and utilizing the electronic device when the sound is in a playing state The first microphone collects the sound of the scene, and the second microphone of the electronic device acquires the sound of the sound playing, wherein the sound of the scene includes the voice command of the user and the sound of the sound playing; the sound and the scene from the scene The voice command of the user is recognized in the voice played by the audio, and the electronic device is operated according to the recognized voice command of the user.

Please refer to FIG. 1. FIG. 1 is a schematic flowchart diagram of a method for voice recognition according to an embodiment of the present invention. As shown in FIG. 1 , a voice recognition method provided by an embodiment of the present invention may include the following content:

101. Acquire an acoustic state of the electronic device.

The electronic device is a smart device with a play function. For example, tablets, smartphones, e-readers, laptops, in-vehicle devices, IPTV, wearables, or other smart devices with playback capabilities.

Wherein, the sound state includes an on/off state of the sound, that is, whether the sound is in a play state.

102. When the audio is in a playing state, the first microphone of the electronic device is used to collect the sound of the scene, and the second microphone of the electronic device is used to obtain the sound of the audio playing, wherein the sound of the scene includes the voice of the user. The command and the sound of the audio playback.

The first microphone is a primary microphone, and the second microphone is a secondary microphone.

The user voice command may perform an operation on the command smart device, for example, instructing the mobile terminal to automatically input text according to the voice of the user, instructing the in-vehicle device to navigate according to the voice of the user, and instructing the mobile phone or the wearable device to answer the question raised by the user. Or other operating instructions.

103. Identify a voice command of the user from the sound of the scene and the sound played by the sound.

Preferably, the electronic device acquires data of the first channel and data of the second channel; the electronic device utilizes a frequency conversion method from data of the first channel and the second channel Obtaining a valid audio data stream in the data; wherein the electronic device performs noise cancellation on the valid audio data stream using an automatic gain control algorithm AGC to obtain a user's voice command.

Wherein the first channel is a left channel; the second channel is a right channel.

Among them, the automatic gain control algorithm AGC is an automatic control algorithm that automatically keeps the output signal within a small range when the input signal changes greatly.

104. The electronic device is operated according to the identified voice instruction of the user.

The operation of the electronic device according to the user voice instruction is obtained by performing voice recognition and semantic analysis on the voice information. The operating the electronic device includes invoking various forms of service information. The service information may be in various media formats, such as text, voice, image, animation, etc.; the mobile terminal may be called from the local storage, or may be obtained by the mobile terminal from the network; The various media information presented to the user in a certain program may also be various media information presented to the user during the running of the program. Specifically, text corresponding to the voice information is obtained by voice recognition. In some applications, such as voice input methods, text can be used as business information. Under normal circumstances, it is also possible to perform semantic analysis on the text, and after semantic analysis, an operation instruction corresponding to the semantics of the text, that is, an operation instruction corresponding to the voice information is obtained. Then, by executing the voice instruction, the service information corresponding to the voice information is obtained.

Optionally, before the operating the electronic device according to the voice instruction of the user, the method further includes: the electronic device outputting the recognized voice command, so that the user confirms the voice command. The output form of the voice command includes a Chinese character output, that is, the user's voice is converted into a text form display, and the user performs the next operation after confirming by clicking. Optionally, the output form of the voice command may also be a voice output. After the electronic device recognizes the voice command, the voice command is output by repeating the voice command, and the user inputs the confirmed voice command or clicks on the screen. The voice command is confirmed.

It can be seen that, in the solution of the embodiment, when the user acquires the sound of the electronic device in the playing state, the digital microphone is used to capture the sound of the scene, and the first microphone is used to collect the sound of the electronic device. The second microphone acquires the sound of the audio playing, wherein the sound of the scene includes the voice command of the user and the sound of the sound playing, and the sound of the live sound acquired by the first microphone according to the sound played by the sound of the second microphone The playback sound is removed, thereby enabling efficient operation of the electronic device by voice commands. This effectively solves the problem of speech recognition technology in complex scenes and greatly improves the user experience.

The electronic device is operated according to the voice instruction of the user. Optionally, the voice command can be identified by using a local voice recognition module, such as a local voice recognition software. Speech recognition module on other remote devices.

Specifically, due to problems such as inaccurate pronunciation or accent, when the terminal recognizes the voice command, the terminal may determine an identification algorithm according to the preset rule, and the recognition algorithm may identify the voice information of the special user, for example, When the user determines the recognition algorithm, the input voice test information is dialect voice information, and the determined recognition algorithm is a recognition algorithm for the dialect voice information of the user. Thereafter, when the voice information to be recognized input by the user is a dialect voice message, The recognition algorithm determined according to the preset rule can identify it, thereby completing the corresponding operation instruction action. When the user needs to perform the corresponding action behavior according to the voice information, the user needs to input the voice information to be recognized to the terminal, and the terminal receives the voice information to be recognized. The voice information to be recognized is identified according to the recognition algorithm. Specifically, since the network or the local voice model library includes at least one user's voice information, and each voice information has a corresponding action behavior, when the terminal receives the voice information to be recognized input by the user, according to the above determination. The identification algorithm identifies the voice information to be recognized input by the user, and when the terminal recognizes that the voice information to be recognized input by the user is the target voice information in the network or the local voice model library, the terminal may correspond to the target voice information. The action behavior identifies the action behavior corresponding to the voice information to be recognized, thereby performing the recognized action behavior.

For example, at present, the terminal generally has a voice photographing function. When the user says “photographing” or “eggplant” to the terminal, the terminal performs voice recognition to determine whether it is a corresponding character, thereby performing a photographing function. However, because the user's pronunciation is not accurate or stuttering, after saying "photograph" or "eggplant" to the terminal, it is still impossible to perform the photographing function, which is relatively embarrassing. At this time, the user can turn on the above-mentioned voice calibration mode, according to the "photographing" provided by the user, etc. The target voice information is used to calibrate the “photograph” or “eggplant” provided by the user, and recognize the voice function that the user wishes to perform, thereby “tailoring” a set of voice recognition system belonging to him. The camera function is implemented.

Through the above technical solutions, it is possible to carry out "tailor-made" speech recognition for special users, and has the advantage of "special individual special treatment", avoids the unified processing, has certain re-learning, and greatly increases the speech recognition rate. , improve the user experience.

Please refer to FIG. 2. FIG. 2 is a schematic flowchart diagram of a method for voice recognition according to another embodiment of the present invention. As shown in FIG. 2, a voice recognition method provided by an embodiment of the present invention may include the following contents:

201. Acquire an audio state of the electronic device.

202. Determine whether the acoustic state is a play state.

If it is in the playing state, step 203 is performed;

If it is in the non-play state, step 208 is performed.

203. When the sound is in the playing state, the first microphone of the electronic device is used to collect the sound of the scene, and the second microphone of the electronic device is used to obtain the sound of the sound playing, wherein the sound of the scene includes the voice command of the user and The sound of the audio playback.

204. The electronic device fills the sound of the scene into the first channel to obtain first audio data.

The electronic device fills the sound of the audio playback into the second channel to obtain second audio data.

205. The electronic device controls, by coordinating the first microphone and the second microphone, a delay of data between the first channel and the second channel to be less than a threshold.

For example, the threshold may be 1 ms, 2 ms, 3 ms, 4 ms, 5 ms, 6 ms, 7 ms, or other values.

Preferably, the threshold is any positive value less than or equal to 2 ms.

It can be understood that in order to achieve good speech recognition, we need to ensure that the sound data stream of the second channel can be continuously received while receiving the sound data stream of the first channel, and the two data streams are guaranteed. Synchronization.

206: Obtain first valid audio data in the first audio data; the electronic device acquires second valid audio data from the second audio data by using a frequency conversion algorithm; and the electronic device uses an automatic gain control algorithm AGC The first valid audio data and the second valid audio data are subjected to noise cancellation processing to identify a user's voice instruction.

207. Perform operations on the electronic device according to the identified voice command of the user.

208. When the audio of the electronic device is in a non-playing state, the electronic device acquires a user voice by using the first microphone and the second microphone; and the electronic device performs noise cancellation on the user voice by using an AGC algorithm. In order to obtain a voice instruction of the user; the electronic device operates according to the voice instruction of the user.

It can be seen that, in the solution of the embodiment, when the user's audio is in the non-playing state, the voice recognition state is converted into a digital double-mike noise reduction technology, and the voice command and the noise are distinguished by the recognition, thereby realizing the instruction.

Please refer to FIG. 3. FIG. 3 is a schematic diagram of an apparatus for voice recognition according to an embodiment of the present invention. As shown in FIG. 3, a voice recognition apparatus provided by an embodiment of the present invention may include the following contents:

The first obtaining unit 301 is configured to acquire a state of the sound of the electronic device;

The first acquiring unit is further configured to: when the sound is in a non-playing state, the electronic device acquires a user voice by using the first microphone and the second microphone;

The collecting unit 302 is configured to collect the sound of the scene by using the first microphone of the electronic device when the sound is in the playing state, and acquire the sound of the sound playing by using the second microphone of the electronic device, wherein the sound of the scene is The voice command of the user and the sound of the audio play are included.

The sound collection device may include a microphone array or a wearable sound collection device in the mobile terminal. The wearable sound collecting device may be an electronic skin tattoo worn on the throat of the user, or a bone sensing microphone worn in the user's cochlea. The mobile terminal can select a sound collecting device for collecting sound according to its own azimuth motion state.

Preferably, the sound collection device is a dual microphone sound collection device.

The second obtaining unit 303 is configured to recognize a voice command of the user from the sound of the scene and the sound played by the sound.

Preferably, the second obtaining unit is specifically configured to: acquire data of the first channel and data of the second channel; and use data from the first channel and the second by using a frequency conversion method Acquiring an effective audio data stream in the data of the channel; performing noise cancellation on the valid audio data stream by using an automatic gain control algorithm AGC to acquire a user's voice command.

The operation unit 304 operates the electronic device according to the recognized voice instruction of the user.

The first obtaining unit 301, the collecting unit 302, the second obtaining unit 303, and the operating unit 304 may be configured to perform the methods described in steps 101, 102, 103, and 104 in Embodiment 1. For details, see Embodiment 1 The description of the method will not be repeated here.

Referring to FIG. 4, FIG. 4 is a schematic diagram of another apparatus for voice recognition according to an embodiment of the present invention. As shown in FIG. 4, a voice recognition apparatus provided by an embodiment of the present invention may include the following contents:

The first obtaining unit 401 is configured to acquire a state of the sound of the electronic device;

The collecting unit 402 is configured to collect the sound of the scene by using the first microphone of the electronic device when the sound is in the playing state, and acquire the sound of the sound playing by using the second microphone of the electronic device, wherein the sound of the scene The voice command of the user and the sound of the audio play are included.

The processing unit 403 is configured to fill the first channel with the sound of the scene to obtain first audio data; the electronic device fills the sound of the audio to the second channel to obtain second audio data.

The control unit 404, the electronic device controls the delay of data between the first channel and the second channel to be less than a threshold by coordinating the first microphone and the second microphone.

For example, the threshold may be 1 ms, 2 ms, 3 ms, 4 ms, 5 ms, 6 ms, 7 ms, or other values.

Preferably, the threshold is any positive value less than or equal to 2 ms.

The second obtaining unit 405 is configured to recognize a voice command of the user from the sound of the scene and the sound played by the sound.

Optionally, the second obtaining unit is further configured to: after the electronic device acquires the voice instruction, output the voice command, so that the user confirms.

Optionally, the second obtaining unit is further configured to: obtain the recognition degree of the voice recognition service information by the user according to the collected confirmation information, and receive the user to the voice recognition service information. Recognition information.

The operation unit 406 operates the electronic device according to the recognized voice instruction of the user.

The first obtaining unit 401, the collecting unit 402, the processing unit 403, the control unit 404, the second obtaining unit 405, and the operating unit 406 may be configured to perform steps 201, 202, 203, 204, 205, and 206 in Embodiment 2. For a detailed description of the method, refer to the description of the method in Embodiment 2, and details are not described herein again.

It can be seen that, in the solution of the embodiment, the device adds the control unit 404 and the processing unit 403, and also increases the confirmation of the recognized voice command, which greatly improves the user experience.

Referring to FIG. 5, FIG. 5 is a schematic diagram of another apparatus for voice recognition according to an embodiment of the present invention. As shown in FIG. 5, a voice recognition apparatus provided by an embodiment of the present invention may include the following contents:

The first obtaining unit 501 is configured to acquire a state of the sound of the electronic device;

The collecting unit 502 is configured to collect the sound of the scene by using the first microphone of the electronic device when the sound is in the playing state, and acquire the sound of the sound playing by using the second microphone of the electronic device, wherein the sound of the scene The voice command of the user and the sound of the audio play are included.

The processing unit 503 is configured to fill the first channel with the sound of the scene to obtain first audio data; and the electronic device fills the sound of the audio to the second channel to obtain second audio data.

The control unit 504, the electronic device, by coordinating the first microphone and the second microphone, delays a data between the first channel and the second channel by less than a threshold.

The second obtaining unit 505 is configured to recognize a voice command of the user from the sound of the scene and the sound of the sound playing.

Optionally, the second obtaining unit is further configured to: after the electronic device acquires the voice instruction, output the voice command, so that the user confirms.

Optionally, the second obtaining unit is further configured to acquire the recognition degree of the voice recognition service information by the user according to the collected confirmation information and the information collection template, and receive the user pair The voice recognition service information has the degree of acceptance information.

The operating unit 506 is configured to operate the electronic device according to the recognized voice command of the user.

The optimization unit 507 is configured to optimize the speech recognition algorithm according to the recognition degree of the user for the voice recognition service information acquired by the acquisition unit 505.

Please refer to FIG. 6 , which is a schematic structural diagram of an electronic device according to an embodiment of the present invention. The electronic device 600 in the embodiment of the present invention may be different types of electronic devices, such as: smart phones, tablet computers, handheld computers, and mobile Internet devices, personal digital assistants, media players, smart televisions, smart watches, smart glasses, and smart phones. Hand ring and so on.

As shown in FIG. 6, the electronic device 600 in the embodiment of the present invention includes: at least one processing device 610, such as a CPU, at least one receiving device 613, at least one storage device 614, at least one transmitting device 615, and at least one communication bus 612. The communication bus 612 is used to implement connection communication between these components. The receiving device 613 and the sending device 615 may be wired transmission ports, or may be wireless devices, for example, including antenna devices, for performing data communication with other devices. The storage device 614 may be a high speed RAM memory or a non-volatile memory such as at least one disk storage.

The processing device 610 can execute an operating system of the electronic device 600 and various installed applications, program codes, and the like, for example, each unit described above, including the first acquiring unit 301 (401, 501), the Acquisition unit 302 (402, 502), second acquisition unit 303 (405, 505), operation unit 304 (406, 506), processing unit 403 (503), control unit 404 (504), optimization Unit 507, etc.

Program code is stored in the storage device 614, and the processing device 610 can invoke program code stored in the storage device 614 via the communication bus 612 to perform related functions. For example, each unit described in FIG. 3, FIG. 4, FIG. 5 (for example, the first acquisition unit 301, the collection unit 302, the second acquisition unit 303, the operation unit 304, etc.) is stored. The program code in the storage device 614 is executed by the processing device 610 to implement the functions of the respective units to implement the process of speech recognition.

In one embodiment of the invention, the storage device 614 stores a plurality of instructions that are executed by the processing device 610 to implement a speech recognition method. Specifically, the processing device 610 acquires the state of the sound of the electronic device; when the sound is in the playing state, the processing device 610 collects the sound of the scene by using the first microphone of the electronic device, and utilizes the electronic device The second microphone acquires the sound of the sound playing, wherein the sound of the scene includes the voice command of the user and the sound of the sound playing; the processing device 610 identifies from the sound of the scene and the sound played by the sound a voice command of the user; the processing device 610 operates the electronic device according to the recognized voice command of the user.

In a further embodiment, the processing device 610 fills in the sound of the scene into the first channel before the processing device 610 identifies the user's voice command from the sound of the scene and the sound played by the sound. Obtaining first audio data; the processing device 610 fills the sound of the audio playback into the second channel to obtain second audio data.

In a further embodiment, the processing device 610 identifies the voice command of the user from the sound of the scene and the sound of the sound, specifically including:

The processing device 610 acquires first valid audio data from the first audio data by using a frequency conversion algorithm;

The processing device 610 acquires second valid audio data from the second audio data by using a frequency conversion algorithm;

The processing device 610 performs noise cancellation processing on the first valid audio data and the second effective audio data by using an automatic gain control algorithm AGC to recognize a user's voice instruction.

In a further embodiment, the processing device 610 controls the delay of data between the first channel and the second channel to be less than a threshold by coordinating the first microphone and the second microphone. In a further embodiment, the processing device 610 acquires user speech using the first microphone and the second microphone when the audio of the electronic device is in a non-playing state; the processing device 610 utilizes an AGC algorithm pair The user voice performs noise cancellation to acquire a voice instruction of the user; the processing device 610 operates according to the voice instruction of the user.

For a specific implementation of the above-mentioned instructions, the processing device 610 may refer to the description of related steps in the corresponding embodiments in FIG. 1 to FIG. 2, and details are not described herein.

It can be seen that, in the solution of the embodiment, the user effectively solves the problem of the voice recognition technology in the complex scenario, and the embodiment further increases the voice recognition recognition degree acquisition and the optimization module for optimizing the voice recognition device according to the approval data. , greatly improving the user experience.

In the above embodiments, the descriptions of the various embodiments are different, and the details that are not detailed in a certain embodiment can be referred to the related descriptions of other embodiments.

In the several embodiments provided herein, it should be understood that the disclosed apparatus may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or may be Integrate into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be electrical or otherwise.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product stored in a storage medium. A number of instructions are included to cause a computer device (which may be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: a U disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk, and the like. .

Claims (10)

A method for speech recognition, characterized in that the method comprises: Obtaining the state of the sound of the electronic device; When the sound is in a playing state, the first microphone of the electronic device is used to collect the sound of the scene, and the second microphone of the electronic device is used to obtain the sound of the sound playing, wherein the sound of the scene includes the voice instruction of the user and The sound of the sound playing; Identifying a user's voice command from the sound of the scene and the sound played by the sound; The electronic device is operated according to the recognized voice instruction of the user. The method according to claim 1, wherein the method further comprises: before the voice instruction of the user is recognized from the sound of the scene and the sound of the sound playing, the method further comprising: The electronic device fills the sound of the scene into the first channel to obtain first audio data; The electronic device fills the sound of the audio playback into the second channel to obtain second audio data. The method according to claim 2, wherein the electronic device identifies the voice command of the user from the sound of the scene and the sound of the sound, specifically comprising: The electronic device acquires first valid audio data from the first audio data by using a frequency conversion algorithm; The electronic device acquires second valid audio data from the second audio data by using a frequency conversion algorithm; The electronic device performs noise cancellation processing on the first valid audio data and the second effective audio data by using an automatic gain control algorithm AGC to identify a user's voice instruction. The method of claim 2, wherein the method further comprises: The electronic device controls the delay of data between the first channel and the second channel to be less than a threshold by coordinating the first microphone and the second microphone. The method of claim 1 further comprising: When the sound of the electronic device is in a non-playing state, the electronic device acquires a user voice by using the first microphone and the second microphone; The electronic device performs noise cancellation on the user voice by using an AGC algorithm to acquire a voice instruction of the user; The electronic device operates in accordance with a voice instruction of the user. A device for speech recognition, characterized in that the device comprises: a first acquiring unit configured to acquire a state of an audio of the electronic device; The collecting unit is configured to collect the sound of the scene by using the first microphone of the electronic device when the sound is in the playing state, and acquire the sound of the sound playing by using the second microphone of the electronic device; The sound of the scene includes a voice command of the user and a sound played by the sound; a second acquiring unit, configured to acquire a voice instruction of the user from the sound of the scene and the sound played by the sound; The operating unit is configured to operate the electronic device according to the recognized voice command of the user. The device of claim 6 further comprising The processing unit is configured to: the electronic device fills the sound of the scene into the first channel to obtain the first audio data, and fills the sound of the audio to the second channel to obtain the second audio data. The device according to claim 7, wherein the second obtaining unit is specifically configured to: Obtaining data of the first channel and data of the second channel; Acquiring a valid audio data stream from the data of the first channel and the data of the second channel by using a frequency conversion method; The effective audio data stream is subjected to noise cancellation using an automatic gain control algorithm AGC to acquire a user's voice command. The device according to claim 6 or 7, wherein the device further comprises: And a control unit configured to: control, by the electronic device, the delay of data between the first channel and the second channel by using the first microphone and the second microphone to be less than a threshold. The device of claim 6 wherein: The first obtaining unit is further configured to: when the sound is in a non-playing state, the electronic device acquires a user voice by using the first microphone and the second microphone; The speech is noise cancelled using an AGC algorithm to obtain the user's voice instructions.

Images