WO2020258570A1 - 语音采集控制方法、装置及tws耳机 - Google Patents

语音采集控制方法、装置及tws耳机 Download PDF

Info

Publication number
WO2020258570A1
WO2020258570A1 PCT/CN2019/109022 CN2019109022W WO2020258570A1 WO 2020258570 A1 WO2020258570 A1 WO 2020258570A1 CN 2019109022 W CN2019109022 W CN 2019109022W WO 2020258570 A1 WO2020258570 A1 WO 2020258570A1
Authority
WO
WIPO (PCT)
Prior art keywords
earphone
noise
slave
mic
headset
Prior art date
Application number
PCT/CN2019/109022
Other languages
English (en)
French (fr)
Inventor
杨宗旭
Original Assignee
歌尔股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 歌尔股份有限公司 filed Critical 歌尔股份有限公司
Priority to US17/608,322 priority Critical patent/US11937055B2/en
Publication of WO2020258570A1 publication Critical patent/WO2020258570A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H04R1/1083—Reduction of ambient noise
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H04R1/1058—Manufacture or assembly
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16—Sound input; Sound output
    • G06F3/165—Management of the audio stream, e.g. setting of volume, audio stream path
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16—Sound input; Sound output
    • G06F3/167—Audio in a user interface, e.g. using voice commands for navigating, audio feedback
    • 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
    • G10L15/08—Speech classification or search
    • 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
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/78—Detection of presence or absence of voice signals
    • G10L25/84—Detection of presence or absence of voice signals for discriminating voice from noise
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H04R1/1041—Mechanical or electronic switches, or control elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/20—Arrangements for obtaining desired frequency or directional characteristics
    • H04R1/32—Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only
    • H04R1/40—Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers
    • H04R1/406—Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers microphones
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R29/00—Monitoring arrangements; Testing arrangements
    • H04R29/004—Monitoring arrangements; Testing arrangements for microphones
    • H04R29/005—Microphone arrays
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R3/00—Circuits for transducers, loudspeakers or microphones
    • 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
    • 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
    • G10L15/08—Speech classification or search
    • G10L2015/088—Word spotting
    • 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
    • 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
    • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering
    • G10L21/0216—Noise filtering characterised by the method used for estimating noise
    • G10L2021/02161—Number of inputs available containing the signal or the noise to be suppressed
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H04R1/1016—Earpieces of the intra-aural type
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R2410/00—Microphones
    • H04R2410/05—Noise reduction with a separate noise microphone
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R2420/00—Details of connection covered by H04R, not provided for in its groups
    • H04R2420/07—Applications of wireless loudspeakers or wireless microphones
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R3/00—Circuits for transducers, loudspeakers or microphones
    • H04R3/005—Circuits for transducers, loudspeakers or microphones for combining the signals of two or more microphones

Definitions

  • the present invention relates to the technical field of voice noise reduction, in particular to a voice collection control method, device and TWS headset.
  • voice assistants have become a standard feature of smart headsets.
  • the headset interacts with the voice assistant of the terminal, the headset is required to collect the user's voice.
  • the main earphone is usually fixedly selected to collect the user's voice.
  • the fixed voice collection control method cannot realize the environmental noise Self-adaptation may cause the voice signal collected by the main headset to include more environmental noise, thereby reducing the quality of the collected voice, increasing noise interference, and reducing user experience.
  • the purpose of the present invention is to provide a voice collection control method, device and TWS headset, which reduces the environmental noise included in the voice signal collected by the headset, improves the voice quality of the collection, reduces noise interference, and improves user experience.
  • the present invention provides a voice collection control method, which is applied to a main headset, and includes:
  • the start command includes the time start point and time interval of noise collection
  • the slave earphone performing noise collection according to the start instruction includes:
  • the slave earphone controls its own MIC to start noise collection from the time starting point of the noise collection, and synchronizes the collected noise data to the master earphone every time interval.
  • the ambient noise is determined according to the collected noise data of the master earphone and the noise data of the slave earphone Before the smaller headphones, it also included:
  • an instruction of keyword detection failure is sent to the slave earphone to control the master earphone and the slave earphone to stop noise detection, stop performing subsequent steps, and end this voice collection.
  • the method further includes:
  • the voice collection right of the headset is switched back to the MIC of the master headset.
  • the method further includes:
  • the collected user voice is processed and sent to the voice assistant of the terminal.
  • the master earphone and the slave earphone both include an external MIC arranged outside the earphone body and an internal MIC arranged inside the earphone body;
  • the step of starting the MIC of the master earphone to perform noise collection and sending a start command to the slave earphone when detecting a user's speech so that the slave earphone controls its own MIC to perform noise collection according to the start instruction includes:
  • the external MIC on the master headset is activated for noise collection and a startup command is sent to the slave headset so that the slave headset controls its own external MIC according to the startup instruction Perform noise collection;
  • the controlling the MIC of the earphone with low noise in the environment to collect user voice includes:
  • the determining the earphone with low environmental noise according to the collected noise data of the master earphone and the noise data of the slave earphone includes:
  • the average loudness of the master earphones in the different frequency bands is compared with the average loudness of the slave earphones respectively, and the earphones with a larger number of frequency bands corresponding to a smaller average loudness are determined as earphones with lower environmental noise.
  • the present invention also provides a voice collection control method, which is applied to a slave headset, including:
  • the start command includes the time start point and time interval of noise collection
  • the controlling the MIC of the slave headset to perform noise collection according to the start instruction includes:
  • the method further includes:
  • a voice collection completion instruction is sent to the main headset to switch the voice collection right of the headset back to the MIC of the main headset.
  • the present invention also provides a voice collection control device applied to the main headset, including:
  • the activation module is configured to activate the MIC of the master headset to perform noise collection and send a activation command to the slave headset when the user's speech is detected, so that the slave headset controls its own MIC to collect noise according to the activation instruction;
  • a noise processing module configured to determine an earphone with a low environmental noise according to the collected noise data of the master earphone and the noise data of the slave earphone;
  • the control module is used to control the MIC of the earphone with low environmental noise to collect user voice.
  • the present invention also provides a voice collection control device, which is applied to a slave headset, and includes:
  • a receiving module configured to receive a start command sent by the main earphone to start the MIC of the main earphone for noise collection when the user is speaking;
  • the control module is configured to control the MIC of the slave headset to perform noise collection according to the start instruction, so that the master headset determines the ambient noise in the environment according to the collected noise data of the master headset and the noise data of the slave headset Smaller earphones, and control the MIC of the earphones with lower environmental noise to collect user voice.
  • the present invention also provides a TWS headset, including:
  • Memory used to store computer programs
  • the processor is used to implement the steps of the voice collection control method described above when the computer program is executed.
  • the present invention provides a voice collection control method, which is applied to a master headset.
  • the master headset detects that a user speaks, it will control its own MIC and the MIC of the slave headset for noise collection, and according to the noise data collected by the master headset and the slave
  • the noise data collected by the earphone determines the earphone with lower environmental noise, that is, the earphone with better sound environment is determined, and then the MIC of the earphone with lower environmental noise is controlled for user voice collection. It can be seen that by selecting headphones with low environmental noise for user voice collection, the adaptation to environmental noise is realized, the environmental noise included in the voice signal collected by the headphones is reduced, the quality of the collected voice is improved, and the noise interference is reduced. Improved user experience.
  • the present invention also provides a voice collection control device and a TWS headset, which have the same beneficial effects as the above voice collection control method.
  • Figure 1 is a flow chart of a voice collection control method provided by the present invention
  • FIG. 2 is a schematic diagram of the structure of a master earphone and a slave earphone provided by the present invention
  • Figure 3 is a schematic diagram of a voice collection control provided by the present invention.
  • Figure 4 is a flowchart of a complete voice collection control method provided by the present invention.
  • FIG. 5 is a flowchart of another voice collection control method provided by the present invention, and the method is applied to a slave headset;
  • Figure 6 is a schematic structural diagram of a voice collection control device provided by the present invention.
  • FIG. 7 is a schematic structural diagram of another voice collection control device provided by the present invention.
  • Fig. 8 is a schematic structural diagram of a TWS headset provided by the present invention.
  • the core of the present invention is to provide a voice collection control method, device and TWS headset, which reduces the environmental noise included in the voice signal collected by the headset, improves the collected voice quality, reduces noise interference, and improves user experience.
  • FIG. 1 is a flowchart of a voice collection control method provided by the present invention. The method is applied to a main headset and includes:
  • S11 When detecting the user's speech, start the MIC of the master headset to perform noise collection and send a start command to the slave headset, so that the slave headset controls its own MIC for noise collection according to the start command;
  • S12 Determine the earphone with less noise in the environment according to the collected noise data of the main earphone and the noise data of the slave earphone;
  • this application does not always select a certain earphone for voice collection, but chooses according to the sound environment where the earphone is located.
  • the environmental noise is small, that is, the earphone on the low-noise side performs voice collection, so as to realize the adaptation to the environmental noise.
  • both the main earphone and the slave earphone are provided with MICs.
  • an internal MIC set inside the earphone and an external MIC set outside the earphone can be set.
  • the quantity can be determined according to the actual situation, and this application is not specifically limited here.
  • a connection is also established between the master headset and the slave headset, which can be a wireless connection, and the wireless connection here can be but not limited to a Bluetooth connection.
  • This application does not specifically limit the selection of the left earphone as the master earphone (correspondingly, the right earphone as the slave earphone) or the right earphone as the master earphone (correspondingly, the left earphone as the slave earphone).
  • the master earphone and the slave earphone start noise collection.
  • the master earphone detects the user's speech, it will start its own MIC for noise collection to obtain noise data; in addition, it also sends a start command to the slave earphone, and the slave earphone controls its own MIC according to the start command after receiving the start command. Perform noise collection to obtain noise data.
  • the master headset can immediately control its own MIC for noise collection when it detects the user’s speech, and at the same time send a start command to the slave headset, whether the slave headset receives the start command immediately after noise collection or after a preset time interval The noise collection is determined according to the start command.
  • the main earphone can first save the noise data to the buffer of the main earphone after collecting the noise data. After the noise data is collected from the earphone, the collected noise data will be sent to the main earphone and saved to the buffer Area. During the collection, in order to improve the accuracy of the subsequent environmental noise, it is better to have the data collected in the same time period when the main earphone and the slave earphone are collecting noise data, so that the noise data in the same time period can be extracted during subsequent use. Data processing improves the accuracy of environmental noise determination by controlling variables.
  • the environmental noise usually does not change suddenly, and the data of different time periods (such as two adjacent time periods) can also be used to determine the environmental noise in the future, but it is still recommended to prefer the same time period The noise data for environmental noise determination.
  • the sound environment of the main earphone can be determined according to the collected noise data of the main earphone, and the sound environment of the slave earphone can be determined according to the collected noise data of the slave earphone, and then the environment noise can be selected.
  • Small earphones, and control earphones with less noise in the environment for subsequent user voice collection Specifically, if the environment where the main earphone is in low noise, the main earphone directly controls the MIC of the main earphone for user voice collection. At this time, the MIC of the slave earphone does not work; if the environment where the slave earphone is noisy, the main earphone Send a switching instruction to the slave headset, and the slave headset controls its MIC to work after receiving the switching instruction. At this time, the MIC of the master headset does not work.
  • the main headset of this application when the main headset of this application detects the user's speech, it will control its own MIC and the MIC of the slave headset to collect noise, and determine that the environmental noise is low based on the noise data collected by the master headset and the noise data collected from the headset
  • the earphone of the user is determined to be the earphone with a better sound environment, and then the MIC of the earphone with the lower environmental noise is controlled for user voice collection. It can be seen that by selecting the earphones with low environmental noise for user voice collection, the adaptation to the environmental noise is realized, and the low-noise side is selected as the sound collecting side, which reduces the environmental noise included in the voice signal collected by the earphone and improves the collection The voice quality reduces noise interference and improves user experience.
  • the start command includes the time start point and time interval of noise collection
  • the headphone controls its own MIC to start noise collection from the start point of the noise collection time, and synchronize the collected noise data to the main headphone at intervals.
  • the start command includes the time starting point and time interval of the noise collection
  • the slave headset receives the start command at the start point of the noise collection Start noise collection, and synchronize the collected noise data to the main earphone at intervals.
  • the time starting point here can be, but not limited to, 10 ms after receiving the start instruction, and the time interval is, for example, 100 ms.
  • the present application does not specifically limit the time starting point and time interval here, and they are determined according to actual conditions.
  • the slave headset after receiving the start command, can also send a command confirmation instruction to the master headset to notify the master headset that it has correctly received the start command.
  • the master headset can start timing after sending the start command to the slave headset, and if the command confirmation instruction sent from the headset is not received within the preset time, an abnormal alarm can be issued.
  • Keyword detection based on the collected noise data of the main earphone
  • a keyword detection failure instruction is sent to the slave earphone to control the master earphone and the slave earphone to stop noise detection, and stop performing subsequent steps to end this voice collection.
  • the main earphone After the main earphone detects the noise data, it will first perform the keyword detection of the noise data, where the keywords here can be, for example, "Siri", “Little Ai”, etc. If a keyword is detected, it means that the subsequent user can speak and needs voice collection. At this time, the master headset will send a keyword detection successful instruction to the slave headset to control the master headset and stop the noise detection from the slave headset, so that the follow-up will be based on the collected The noise data of the master earphone and the noise data of the slave earphone determine the earphone with the less noise in the environment.
  • the master headset sends a keyword detection failure instruction to the slave headset to control the master headset and stop noise detection from the slave headset , And stop following steps to end this voice collection.
  • the master headset and the slave headset can stop noise detection at the same time when they stop noise detection, so as to ensure the synchronization of collecting noise data.
  • this way can prevent subsequent user voice collection from being falsely triggered, and improve the reliability of user voice control.
  • this embodiment can be combined with the user's voice assistant to improve the interaction reliability between the headset and the terminal's voice assistant.
  • the method further includes:
  • the voice collection right of the headset is switched back to the MIC of the main headset.
  • the master earphone when the earphone with low noise in the environment is a slave earphone, the master earphone will switch the subsequent user voice collection right to the slave earphone, that is, the MIC of the master earphone does not work, and the MIC of the earphone turns on the user voice Collection, the slave headset sends a voice collection completion instruction to the main headset after the user's voice collection is completed. After the main headset receives the voice collection instruction, it will return to the default state, that is, the MIC of the main headset will detect whether the user is speaking and proceed. Next steps.
  • the slave headset can automatically turn off its own MIC after the user’s voice collection is over, or after sending the voice collection command to the master headset, the master headset controls the slave headset to turn off the MIC of the slave headset.
  • This method is not particularly limited, and is determined according to the actual situation.
  • the user's voice when the user's voice is collected from the headset, it can be set to determine that the current voice collection is completed when the user's voice is not collected for a continuous preset time (such as 1s or 2s), or when the end keyword (such as "end") is collected It is judged that the voice collection is completed this time.
  • a continuous preset time such as 1s or 2s
  • end keyword such as "end”
  • the main earphone judges the environmental noise and switches between the main and slave earphones according to the magnitude of the environmental noise.
  • the slave earphone will automatically switch back to the default state after collecting the user's voice.
  • the control logic is simple and reduces Early development time and cost.
  • the method further includes:
  • the collected user voice is processed and sent to the voice assistant of the terminal.
  • the voice collection control method is combined with the voice assistant of the terminal. Specifically, after the earphone with low noise in the environment collects the user’s voice, the user’s voice is processed.
  • the processing here includes The voice is subjected to beam-forming processing and noise reduction processing, and then sent to the voice assistant module in the headset, and the voice assistant module is transmitted to the voice assistant of the terminal through the communication module in the headset. It can be seen that this implementation combines the voice collection control method with the voice assistant, and improves the reliability of the voice assistant by improving the high quality of the user's voice.
  • the main headset that processes the collected user voice and sends it to the voice assistant of the terminal (if the user voice is collected from the headset, the collected user voice can be sent from the headset to the main headset, which is processed by the main headset
  • the voice assistant that processes the user’s voice directly from the headset and transmits it to the terminal after the user’s voice is collected from the headset.
  • This application is not specifically limited here, and it is based on actual conditions. set.
  • both the master earphone and the slave earphone include an external MIC arranged outside the earphone body and an internal MIC arranged inside the earphone body;
  • start the MIC of the main headset When the user is detected to speak, start the MIC of the main headset to collect noise and send a start command to the slave headset, so that the slave headset controls its own MIC for noise collection according to the start command, including:
  • the external MIC on the main earphone is started for noise collection and a start command is sent to the slave earphone, so that the slave earphone can control its own external MIC for noise collection according to the start command;
  • Control the MIC of the earphone with low noise in the environment to collect user voice including:
  • FIG. 2 is a schematic diagram of the structure of a master earphone and a slave earphone provided by the present invention
  • FIG. 3 is a schematic diagram of a voice collection control provided by the present invention.
  • FIG. 2 there are two external MICs, which are respectively arranged on the earphone column and the rear shell of the earphone head, and the internal MIC is arranged inside the earphone head.
  • this application does not specifically limit the specific number of internal MICs and external MICs, which are determined according to actual conditions.
  • the internal MIC can be used to detect it. Specifically, when the user speaks, the bones of the oral cavity will vibrate. The internal MIC can detect the user’s speech through skeletal vibration, rather than other users speaking, which can be avoided. False triggering improves the reliability of voice collection.
  • the internal MIC will transmit the collected data to the VAD (Voice Activity Detection) module.
  • the VAD module determines that the user speaks based on the data collected by the internal MIC, it sends instructions to the processor of the main headset, and the processor receives After the instruction, turn on the external MIC so that the external MIC can collect noise.
  • the internal MIC here can be a bone conduction sensor, an acceleration sensor, etc., which is not particularly limited in this application.
  • This application considers that after detecting the user’s speech, when performing noise detection, on the one hand, because the noise outside the earphone is collected, on the other hand, when the earphone is well isolated from the outside world, the inner earphone actually collects the external noise.
  • the user’s voice is very small or even impossible to collect. Therefore, in this embodiment, the external MIC of the main earphone and the slave earphone are controlled to collect noise, and the subsequent user voice collection is also collected through the external MIC, thereby improving the noise data and user voice The accuracy of data collection.
  • FIG. 4 is a flowchart of a complete voice collection control method provided by the present invention.
  • S401 The inner ear MIC of the main headset detects the user's speech
  • VAD module trigger
  • S403 Turn on the external MIC of the main earphone to collect sound and perform keyword detection
  • S404 Synchronize noise collection from the external MIC of the headset, and synchronize to the main headset in real time;
  • S405 Judge whether the keyword is detected, if yes, go to S406, otherwise, go to S407;
  • S407 The master earphone and slave earphone stop noise collection synchronously, and enter S400;
  • S409 Determine whether the environmental noise of the main earphone is low, if yes, go to S410, otherwise, go to S412;
  • S411 The user voice is sent to the terminal after beam-forming, noise reduction, and voice assistant modules, and the entry is 415;
  • S412 Perform MIC switching, and collect user voice from the external MIC of the headset;
  • S413 The user's voice passes through the beam-forming, noise reduction, and voice assistant modules, and is sent to the terminal;
  • the internal MIC is the ANC feedback MIC.
  • the internal MIC can also feed back the MIC for the ANC (Active Noise Control) existing in some earphones, thereby realizing the internal MIC and ANC feedback.
  • ANC Active Noise Control
  • the multiplexing of MIC reduces the space occupation of the headset and reduces the cost.
  • the earphone with less noise in the environment is determined, including:
  • the average loudness of the master earphones in different frequency bands is compared with the average loudness of the slave earphones, and the earphones with a larger number of frequency bands corresponding to a smaller average loudness are determined as earphones with lower environmental noise.
  • the processor of the master earphone obtains the noise data of the master earphone and the noise data of the slave earphone in the same time period.
  • Noise data usually includes multi-band speech, which is more accurate in the subsequent determination of environmental noise).
  • Average loudness, and the average loudness of the noise data of the slave headphones in the above different frequency bands (such as low frequency, mid frequency, high frequency), and compare the average loudness of the master headphones in different frequency bands with the average loudness of the slave headphones, that is, the master headphones
  • the average loudness of the low frequency band is compared with the average loudness of the low frequency band of the slave earphone.
  • the average loudness of the middle frequency band of the main earphone is compared with the average loudness of the middle frequency band of the slave earphone.
  • the average loudness of the high frequency band of the master earphone is compared with that of the slave earphone.
  • the average loudness of the high frequency band of the earphones is compared, and it is determined that earphones with a large number of frequency bands (for example, 2 or 3) corresponding to a smaller average loudness are used as earphones with lower environmental noise. Specifically, for example, if the average loudness of 2 or 3 frequency bands of the main earphone is smaller than the average loudness of the frequency band of the slave earphone, it means that the environment noise of the main earphone is low at this time, and the main earphone is selected as the environment where the noise is low. Headphones.
  • the average loudness of several frequency bands of the master headset is equal to the average loudness of the slave headset. If it occurs, it means that the ambient noise of the master headset and the slave headset are the same.
  • which earphone can be selected for subsequent user voice collection for example, by default, the main earphone is selected for collection at this time (there is no need for the main earphone to send instructions to the slave earphone for switching, which simplifies the control logic).
  • control method is simple, the processing time is less, and the real-time performance of earphone voice collection is improved.
  • the noise data collected by the main earphone can also be fitted with the least square method to obtain the frequency response curve (the horizontal axis is the frequency, the vertical axis is the loudness), and the noise data collected from the earphone The least squares method is fitted to obtain the frequency response curve.
  • the environmental noise can be determined, and then Make sure to use headphones with low noise in the environment.
  • other methods can also be used to determine the noise, which is not specifically limited in this application.
  • FIG. 5 is a flowchart of another voice collection control method provided by the present invention.
  • the method is applied to a slave headset and includes:
  • S51 Receive the start command sent by the main earphone to start the MIC of the main earphone for noise collection when the user is speaking;
  • S52 Control the MIC of the slave earphone to collect noise according to the start instruction, so that the master earphone determines the earphone with less noise in the environment according to the collected noise data of the master earphone and the noise data of the slave earphone, and controls the environment where the noise is lower
  • the MIC of the headset performs user voice collection.
  • the start command includes the time start point and time interval of noise collection
  • Control noise collection from the MIC of the headset according to the start command including:
  • the method further includes:
  • the voice collection complete command is sent to the main headset to switch the voice collection right of the headset back to the MIC of the main headset.
  • This embodiment is an embodiment of a voice collection control method for a slave headset paired with a master headset.
  • the voice collection control method applied to the slave headset please refer to the foregoing embodiment, and the present invention will not be repeated here.
  • FIG. 6 is a schematic structural diagram of a voice collection control device provided by the present invention, applied to a main headset, and the device includes:
  • the start module 61 is used to start the MIC of the main earphone for noise collection and send a start command to the slave earphone when the user's speech is detected, so that the slave earphone can control its own MIC for noise collection according to the start command;
  • the noise processing module 62 is configured to determine the earphones with low environmental noise according to the collected noise data of the master earphone and the noise data of the slave earphones;
  • the control module 63 is used to control the MIC of the earphone with low noise in the environment to collect user voice.
  • This embodiment corresponds to the device embodiment of the voice collection control method of the main earphone.
  • the voice collection control device applied to the main earphone please refer to the foregoing embodiment, and the present invention will not be repeated here.
  • FIG. 7 is a schematic structural diagram of another voice collection control device provided by the present invention, which is applied to a slave headset, and the device includes:
  • the receiving module 71 is configured to receive a start command sent by the main earphone to start the MIC of the main earphone for noise collection when the user is speaking;
  • the control module 72 is used to control the MIC of the slave earphone to perform noise collection according to the start instruction, so that the master earphone determines the earphone with less noise in the environment according to the collected noise data of the master earphone and the noise data of the slave earphone, and controls where it is The MIC of the earphone with low environmental noise collects the user's voice.
  • This embodiment is a device embodiment corresponding to the voice collection control method of the slave earphone paired with the master earphone.
  • the voice collection control device applied to the slave earphone please refer to the above embodiment.
  • the present invention is not here. Repeat it again.
  • the present invention also provides a schematic structural diagram of a main earphone, which includes:
  • Memory used to store computer programs
  • the processor is used to implement the steps of the voice collection control method applied to the main earphone when the computer program is executed.
  • the voice collection control method of the main earphone in this embodiment corresponds to the embodiment of the main earphone.
  • the main earphone provided in this embodiment, please refer to the foregoing embodiment, and the present invention will not be repeated here.
  • the present invention also provides a slave headset, including:
  • Memory used to store computer programs
  • the processor is used to implement the steps of the voice collection control method applied to the earphone when the computer program is executed.
  • This embodiment is a method for controlling the voice collection of a slave earphone corresponding to the embodiment of the slave earphone.
  • the slave earphone provided in this embodiment, please refer to the foregoing embodiment, and the present invention will not be repeated here.
  • FIG. 8 is a schematic structural diagram of a TWS headset provided by the present invention.
  • the TWS headset includes:
  • the memory 83 is used for storing computer programs
  • the processor 82 is used to implement the steps of the voice collection control method described above when executing a computer program.
  • TWS includes a master headset and a slave headset.
  • TWS Truste Wireless

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Otolaryngology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Telephone Function (AREA)
  • Headphones And Earphones (AREA)
  • Soundproofing, Sound Blocking, And Sound Damping (AREA)

Abstract

本发明公开了一种语音采集控制方法，应用于主耳机，主耳机在检测到用户说话时，会控制自身的MIC以及从耳机的MIC进行噪声采集，并根据主耳机采集到的噪声数据和从耳机采集到的噪声数据确定环境噪声较小的耳机，也即确定所处声音环境较好的耳机，进而控制环境噪声较小的耳机的MIC进行用户语音采集。可见，通过选取所处环境噪声较小的耳机进行用户语音采集实现了对环境噪声的自适应，减少了耳机采集的语音信号中包括的环境噪声，提高了采集的语音质量，降低了噪声干扰，提升了用户体验。本发明还公开了一种语音采集控制装置及TWS耳机，具有与上述语音采集控制方法相同的有益效果。

Description

语音采集控制方法、装置及TWS耳机
本申请要求于2019年6月28日提交中国专利局、申请号201910576125.1、申请名称为“语音采集控制方法、装置及TWS耳机”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及语音降噪技术领域，特别是涉及一种语音采集控制方法、装置及TWS耳机。
背景技术
随着科技的进步，网络和云服务的不断升级，智能化需求已经成为耳机的重要特征。随着苹果、谷歌、百度等厂商的语音助手平台的相继发布，语音助手已经成为智能耳机的标配功能。在耳机与终端的语音助手交互时，需要耳机对用户的语音进行采集。现有技术中通常是固定选取主耳机来对用户的语音进行采集，但很多时候主耳机和从耳机所处的声音环境是不同的，采用该种固定语音采集控制的方式无法实现对环境噪声的自适应，可能会导致主耳机采集的语音信号中包括的环境噪声较多，从而降低了采集到的语音质量，增大了噪声干扰，降低了用户体验。
发明内容
本发明的目的是提供一种语音采集控制方法、装置及TWS耳机，减少了耳机采集的语音信号中包括的环境噪声，提高了采集的语音质量，降低了噪声干扰，提升了用户体验。
为解决上述技术问题，本发明提供了一种语音采集控制方法，应用于主耳机，包括：
在检测到用户说话时，启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的MIC进行噪声采集；
根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机；
控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
优选地，所述启动命令包括噪声采集的时间起始点和时间间隔；
所述从耳机根据所述启动指令进行噪声采集，包括：
所述从耳机控制自身的MIC从所述噪声采集的时间起始点开始噪声采集，并每隔所述时间间隔将采集到的噪声数据同步至所述主耳机。
优选地，所述启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机之后，所述根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机之前，还包括：
根据采集到的所述主耳机的噪声数据进行关键词检测；
若检测到关键词，则发送关键词检测成功的指令至所述从耳机，以控制所述主耳机和所述从耳机停止噪声检测，并进行后续步骤；
若没有检测到关键词，则发送关键词检测失败的指令至所述从耳机，以控制所述主耳机和所述从耳机停止噪声检测，并停止进行后续步骤，结束本次语音采集。
优选地，在所述所处环境噪声较小的耳机为从耳机时，还包括：
接收所述从耳机在用户语音采集结束后发送的语音采集完毕指令后，将耳机的语音采集权切回至所述主耳机的MIC。
优选地，所述控制所述所处环境噪声较小的耳机的MIC进行用户语音采集之后，还包括：
将采集到的用户语音进行处理后发送至终端的语音助手。
优选地，所述主耳机和所述从耳机均包括设置于耳机本体外部的外部MIC和设置于耳机本体内部的内部MIC；
所述在检测到用户说话时，启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的MIC进行噪声采集，包括：
在通过所述主耳机的内部MIC检测到用户说话时，启动所述主耳机上的外部MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的外部MIC进行噪声采集；
所述控制所述所处环境噪声较小的耳机的MIC进行用户语音采集，包括：
控制所述所处环境噪声较小的耳机的外部MIC进行用户语音采集。
优选地，所述根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机，包括：
获取相同时间段内的所述主耳机的噪声数据和所述从耳机的噪声数据；
根据所述时间段内的所述主耳机的噪声数据得到所述主耳机在不同频段对应的平均响度以及根据所述时间段内的所述从耳机的噪声数据得到与所述主耳机在相同频段的所述从耳机的平均响度；
分别对所述不同频段的主耳机的平均响度与从耳机的平均响度进行比较，确定平均响度较小对应的频段数多的耳机作为所处环境噪声较小的耳机。
为解决上述技术问题，本发明还提供了一种语音采集控制方法，应用于从耳机，包括：
接收主耳机在检测到用户说话时启动所述主耳机的MIC进行噪声采集并发送的启动命令；
根据所述启动指令控制所述从耳机的MIC进行噪声采集，以便所述主耳机根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机，并控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
优选地，所述启动命令包括噪声采集的时间起始点和时间间隔；
所述根据所述启动指令控制所述从耳机的MIC进行噪声采集，包括：
控制所述从耳机的MIC从所述噪声采集的时间起始点开始噪声采集，并每隔所述时间间隔将采集到的噪声数据同步至所述主耳机。
优选地，在所述所处环境噪声较小的耳机为从耳机时，还包括：
在用户语音采集结束后，发送语音采集完毕指令至所述主耳机，以将耳机的语音采集权切回至所述主耳机的MIC。
为解决上述技术问题，本发明还提供了一种语音采集控制装置，应用于主耳机，包括：
启动模块，用于在检测到用户说话时，启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的MIC进行噪声采集；
噪声处理模块，用于根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机；
控制模块，用于控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
为解决上述技术问题，本发明还提供了一种语音采集控制装置，应用于从耳机，包括：
接收模块，用于接收主耳机在检测到用户说话时启动所述主耳机的MIC进行噪声采集并发送的启动命令；
控制模块，用于根据所述启动指令控制所述从耳机的MIC进行噪声采集，以便所述主耳机根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机，并控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
为解决上述技术问题，本发明还提供了一种TWS耳机，包括：
MIC；
存储器，用于存储计算机程序；
处理器，用于执行所述计算机程序时实现如上述所述语音采集控制方法的步骤。
本发明提供了一种语音采集控制方法，应用于主耳机，主耳机在检测到用户说话时，会控制自身的MIC以及从耳机的MIC进行噪声采集，并根据主耳机采集到的噪声数据和从耳机采集到的噪声数据确定环境噪声较小的耳机，也即确定所处声音环境较好的耳机，进而控制环境噪声较小的耳机的MIC进行用户语音采集。可见，通过选取所处环境噪声较 小的耳机进行用户语音采集实现了对环境噪声的自适应，减少了耳机采集的语音信号中包括的环境噪声，提高了采集的语音质量，降低了噪声干扰，提升了用户体验。
本发明还提供了一种语音采集控制装置及TWS耳机，具有与上述语音采集控制方法相同的有益效果。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对现有技术和实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本发明提供的一种语音采集控制方法的流程图；
图2为本发明提供的一种主耳机和从耳机的结构示意图；
图3为本发明提供的一种语音采集控制的原理图；
图4为本发明提供的一种完整的语音采集控制方法的流程图；
图5为本发明提供的另一种语音采集控制方法的流程图，该方法应用于从耳机；
图6为本发明提供的一种语音采集控制装置的结构示意图；
图7为本发明提供的另一种语音采集控制装置的结构示意图；
图8为本发明提供的一种TWS耳机的结构示意图。
具体实施方式
本发明的核心是提供一种语音采集控制方法、装置及TWS耳机，减少了耳机采集的语音信号中包括的环境噪声，提高了采集的语音质量，降低了噪声干扰，提升了用户体验。
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
请参照图1，图1为本发明提供的一种语音采集控制方法的流程图，该方法应用于主耳机，包括：
S11：在检测到用户说话时，启动主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便从耳机根据启动指令控制自身的MIC进行噪声采集；
S12：根据采集到的主耳机的噪声数据和从耳机的噪声数据确定所处环境噪声较小的耳机；
S13：控制所处环境噪声较小的耳机的MIC进行用户语音采集。
与现有技术中固定选取主耳机来对用户的语音进行采集不同的是，本申请并不固定选取某一耳机来进行语音采集，而是根据耳机所处的声音环境来选择，具体选择所处环境噪声较小也即低噪声侧的耳机来进行语音采集，从而实现对环境噪声的自适应。
首先需要说明的是，本申请中，主耳机和从耳机上均设置有MIC，例如具体可设置有设置于耳机内部的内部MIC和设置于耳机外部的外部MIC，对于内部MIC和外部MIC的具体数量则可根据实际情况来定，本申请在此不作特别的限定。此外，主耳机和从耳机之间还建立连接，具体可以为无线连接，且这里的无线连接可以但不仅限为蓝牙连接。本申请对于具体选用左耳机作为主耳机(相应地，右耳机作为从耳机)还是右耳机作为主耳机(相应地，左耳机作为从耳机)不作特别的限定。
具体地，站在功耗的角度，为减少功耗，同时也为减少处理器的负担，只有在检测到用户说话时，主耳机和从耳机才开启噪声采集。具体地，主耳机在检测到用户说话时，会启动自身的MIC进行噪声采集，得到噪声数据；此外，还向从耳机发送启动命令，从耳机在接收到启动命令后按照启动命令控制自身的MIC进行噪声采集，得到噪声数据。在实际应用中，主耳机在检测到用户说话时可以立即控制自身的MIC进行噪声采集，同时向从耳机发送启动命令，从耳机在接收到启动命令后是立即进行噪声采集还是间隔预设时间后再进行噪声采集根据启动命令来确定。
在采集过程中，主耳机在采集到噪声数据后可以先将噪声数据保存至主耳机的缓冲区，从耳机在采集到噪声数据后会将采集到的噪声数据发送至主耳机，并保存至缓冲区。在采集时，为了提高后续环境噪声的精度，主耳机和从耳机在进行噪声数据采集时最好有在相同时间段采集的数据，以便在后续使用过程中提取出相同时间段内的噪声数据进行数据处理，通过控制变量提高环境噪声确定精度。当然，考虑到采集时间不会特别长，环境噪声通常不会发生突变，后续也可利用不同时间段(例如相邻两个时间段)的数据进行环境噪声确定，但还是推荐优先选用相同时间段的噪声数据进行环境噪声确定。
后续在使用过程中，可以根据采集到的主耳机的噪声数据确定主耳机所处的声音环境，根据采集到的从耳机的噪声数据确定从耳机所处的声音环境，进而选取出所处环境噪声较小的耳机，并控制所处环境噪声较小的耳机进行后续的用户语音采集。具体地，如果主耳机所处环境噪声较小，则主耳机直接控制主耳机的MIC进行用户语音采集，此时，从耳机的MIC不工作；如果从耳机所处环境噪声较小，则主耳机发送切换指令至从耳机，从耳机在接收到切换指令后控制其MIC工作，此时主耳机的MIC不工作。
还需要说明的是，S11-S13的整个过程时间非常短，例如2s、1s甚至更短(但为了保证环境噪声确定精度，采集时间也不能太短)，对后续用户的语音采集影响非常小甚至完全不影响。
综上，本申请主耳机在检测到用户说话时，会控制自身的MIC以及从耳机的MIC进行噪声采集，并根据主耳机采集到的噪声数据和从耳机采集到的噪声数据确定环境噪声较小的耳机，也即确定所处声音环境较好的耳机，进而控制环境噪声较小的耳机的MIC进行用户语音采集。可见，通过选取所处环境噪声较小的耳机进行用户语音采集实现了对环境噪声的自适应，选取低噪声侧作为采音侧，减少了耳机采集的语音信号中包括的环境噪声，提高了采集的语音质量，降低了噪声干扰，提升了用户体验。
在上述实施例的基础上：
作为一种优选地实施例，启动命令包括噪声采集的时间起始点和时间间隔；
从耳机根据启动指令进行噪声采集，包括：
从耳机控制自身的MIC从噪声采集的时间起始点开始噪声采集，并每隔时间间隔将采集到的噪声数据同步至主耳机。
主耳机在通过发送启动命令以控制从耳机根据启动命令进行噪声采集时，启动命令中包括噪声采集的时间起始点和时间间隔，则从耳机在接收到启动命令后，会在噪声采集的起始点开始进行噪声采集，并每隔时间间隔将采集的噪声数据同步至主耳机。具体地，这里的时间起始点可以但不仅限为接到启动指令后10ms，时间间隔例如为100ms，本申请对于这里的时间起始点和时间间隔不作特别的限定，根据实际情况来定。
此外，在实际应用中，从耳机在接收到启动命令后，还可以发送命令确认指令至主耳机，以通知主耳机其已经正确接收到启动命令。主耳机在发送启动命令至从耳机后可以开始计时，如果在预设时间内没有接收到从耳机发送的命令确认指令，则可发出异常告警。
作为一种优选地实施例，启动主耳机的MIC进行噪声采集并发送启动命令至从耳机之后，根据采集到的主耳机的噪声数据和从耳机的噪声数据确定所处环境噪声较小的耳机之前，还包括：
根据采集到的主耳机的噪声数据进行关键词检测；
若检测到关键词，则发送关键词检测成功的指令至从耳机，以控制主耳机和从耳机停止噪声检测，并进行后续步骤；
若没有检测到关键词，则发送关键词检测失败的指令至从耳机，以控制主耳机和从耳机停止噪声检测，并停止进行后续步骤，结束本次语音采集。
为了尽量避免误触发，本实施例中，主耳机在检测到噪声数据后，会先进行噪声数据的关键词检测，其中，这里的关键词例如可以为“Siri”、“小爱同学”等，如果检测到关键词，则说明后续用户会说话且需要语音采集，此时主耳机会发送关键词检测成功的指令至从耳机，以控制主耳机和从耳机停止噪声检测，以便后续根据采集到的主耳机的噪声数据和从耳机的噪声数据确定所处环境噪声较小的耳机。如果没有检测到关键词，则说明用户 虽然要说话，但此时并不需要进行语音采集，则此时主耳机发送关键词检测失败的指令至从耳机，以控制主耳机和从耳机停止噪声检测，并停止进行后续步骤，结束本次语音采集。此外，不管有没有检测到关键词，主耳机和从耳机在停止噪声检测时均可以同时停止噪声检测，以保证采集噪声数据的同步。当然，也可以不同步停止噪声检测，本申请在此不做特别地限定。
可见，通过该种方式可以避免后续的用户语音采集被误触发，提高了用户语音控制的可靠性。在实际应用中，本实施例可以与用户语音助手相结合，提高了耳机与终端的语音助手的交互可靠性。
作为一种优选地实施例，在所处环境噪声较小的耳机为从耳机时，还包括：
接收从耳机在用户语音采集结束后发送的语音采集完毕指令后，将耳机的语音采集权切回至主耳机的MIC。
具体地，在所处环境噪声较小的耳机为从耳机时，则主耳机会将后续用户语音的采集权切至从耳机，也即主耳机的MIC不工作，从耳机的MIC开启用户语音的采集，从耳机在用户语音采集完毕后向主耳机发送语音采集完毕指令，主耳机在接收到语音采集完毕指令后，会重新回到默认状态，也即主耳机的MIC来检测用户是否说话并进行后续步骤。具体地，从耳机可以在用户语音采集结束后自动关闭自身的MIC，也可以是向主耳机发送完语音采集完毕指令后，由主耳机控制从耳机关闭从耳机的MIC，本申请对于具体选用哪种方式不作特别的限定，根据实际情况来定。
此外，从耳机在进行用户语音采集时，可以设置当连续预设时长(例如1s或者2s)未采集到用户语音时判定本次语音采集完毕，或者当采集到结束关键词(例如“结束”)时判定本次语音采集完毕，当然，这里还可以是其他方式，本申请在此不作特别的限定。
可见，本实施例中，均由主耳机进行环境噪声的判断以及根据环境噪声的大小进行主从耳机的切换，从耳机在采集完用户语音后会自动切回默认状态，控制逻辑简单，减少了前期开发时间和成本。
作为一种优选地实施例，控制所处环境噪声较小的耳机的MIC进行用户语音采集之后，还包括：
将采集到的用户语音进行处理后发送至终端的语音助手。
本实施例中，将语音采集控制方法与终端的语音助手相结合，具体地，在所处环境噪声较小的耳机采集到用户语音后，对用户语音进行处理，其中，这里的处理包括对用户语音进行Beam-forming处理、降噪处理，再送至耳机中的语音助手模块，语音助手模块通过耳机中的通信模块传送至终端的语音助手。可见，本实施将语音采集控制方法与语音助手相结合，通过提高用户语音的高质量，提高了语音助手的使用可靠性。
当然，这里可以是主耳机将采集到的用户语音进行处理后发送至终端的语音助手(如果是从耳机进行用户语音采集，从耳机可以将采集到的用户语音发送至主耳机，由主耳机 处理后传送至终端的语音助手)，也可以是从耳机在采集到用户语音后，从耳机直接对用户语音进行处理并传送至终端的语音助手，本申请在此不作特别的限定，根据实际情况来定。
作为一种优选地实施例，主耳机和从耳机均包括设置于耳机本体外部的外部MIC和设置于耳机本体内部的内部MIC；
在检测到用户说话时，启动主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便从耳机根据启动指令控制自身的MIC进行噪声采集，包括：
在通过主耳机的内部MIC检测到用户说话时，启动主耳机上的外部MIC进行噪声采集并发送启动命令至从耳机，以便从耳机根据启动指令控制自身的外部MIC进行噪声采集；
控制所处环境噪声较小的耳机的MIC进行用户语音采集，包括：
控制所处环境噪声较小的耳机的外部MIC进行用户语音采集。
请参照图2及图3，其中，图2为本发明提供的一种主耳机和从耳机的结构示意图，图3为本发明提供的一种语音采集控制的原理图。
具体地，图2中，外部MIC为2个，分别设置在耳机柱和耳机头的后壳上，内部MIC设置于耳机头的内部。当然，本申请对于内部MIC和外部MIC的具体数量不作特别的限定，根据实际情况来定。
在初期检测用户说话时，可以通过内部MIC来进行检测，具体地，当用户说话时，口腔骨骼会振动，内部MIC通过骨骼振动能够检测到用户说话，而并非是别的用户说话，从而可以避免误触发，提高了语音采集的可靠性。内部MIC会将采集到的数据传送至VAD(Voice Activity Detection，语音活动检测)模块，VAD模块根据内部MIC采集到的数据判定用户说话时，发送指令至主耳机的处理器，处理器在接收到指令后开启外部MIC，以便外部MIC进行噪声采集。这里的内部MIC可以骨传导传感器、加速度传感器等，本申请在此不作特别的限定。
本申请考虑到在检测到用户说话以后，在进行噪声检测时，一方面，由于采集的是耳机外部的噪声，另一方面，在耳机与外界隔离较好时，内耳机实际上采集到的外部用户说话声很小甚至采集不到，因此，本实施例中，控制主耳机和从耳机的外部MIC进行噪声采集，后续的用户语音采集也通过外部MIC进行采集，从而提高了噪声数据以及用户语音数据的采集精度。
请参照图4，图4为本发明提供的一种完整的语音采集控制方法的流程图。
S400：开始
S401：主耳机的内耳MIC检测到用户说话；
S402：VAD模块触发；
S403：开启主耳机的外部MIC采音，进行关键词检测；
S404：同步从耳机的外部MIC进行噪声采集，并实时同步至主耳机；
S405：判断是否检测到关键词，如果是，进入S406，否则，进入S407；
S406：主耳机和从耳机同步停止噪声采集；
S407：主耳机和从耳机同步停止噪声采集，并进入S400；
S408：主耳机判定双耳所处环境噪声；
S409：判断主耳机所处环境噪声是否较小，如果是，进入S410，否则，进入S412；
S410：主耳机的外部MIC采集用户语音；
S411：用户语音经过Beam-forming、降噪以及语音助手模块，并发送至终端，进入是415；
S412：进行MIC切换，从耳机的外部MIC采集用户语音；
S413：用户语音经过Beam-forming、降噪以及语音助手模块，并发送至终端；
S414：语音采集完毕后，切换MIC至主耳机，进入S415；
S415：结束。
作为一种优选地实施例，内部MIC为ANC反馈MIC。
具体地，内部MIC除了可以为上述提到的骨传导传感器、加速度传感器外，还可以为一些耳机中已有的ANC(Active Noise Control，主动降噪)反馈MIC，从而实现了内部MIC与ANC反馈MIC的复用，减小了对耳机的空间的占用，降低了成本。
作为一种优选地实施例，根据采集到的主耳机的噪声数据和从耳机的噪声数据确定所处环境噪声较小的耳机，包括：
获取相同时间段内的主耳机的噪声数据和从耳机的噪声数据；
根据时间段内的主耳机的噪声数据得到主耳机在不同频段对应的平均响度以及根据时间段内的从耳机的噪声数据得到与主耳机在相同频段的从耳机的平均响度；
分别对不同频段的主耳机的平均响度与从耳机的平均响度进行比较，确定平均响度较小对应的频段数多的耳机作为所处环境噪声较小的耳机。
具体地，在主耳机和从耳机采集到噪声数据后，主耳机的处理器获取相同时间段内的主耳机的噪声数据和从耳机的噪声数据(这里之所以选取一段时间是考虑到一段时间内噪声数据中通常包括多频段语音，在后续确定环境噪声时更准确)，对主从耳机的噪声数据进行频响分析，得到主耳机的噪声数据在不同频段(例如低频、中频、高频)的平均响度，以及从耳机的噪声数据在上述不同频段(例如低频、中频、高频)的平均响度，并分别对不同频段的主耳机的平均响度与从耳机的平均响度进行比较，也即主耳机的低频段的平均 响度与从耳机的低频段的平均响度进行大小比较，主耳机的中频段的平均响度与从耳机的中频段的平均响度进行大小比较，主耳机的高频段的平均响度与从耳机的高频段的平均响度进行大小比较，确定平均响度较小对应的频段数多(例如2个或者3个)的耳机作为所处环境噪声较小的耳机。具体地，例如主耳机存在2个或者3个频段的平均响度小于从耳机的频段的平均响度，则说明此时主耳机所处环境噪声较小，此时选取主耳机作为所处环境噪声较小的耳机。
需要说明的是，在实际应用中，很少会出现主耳机的几个频段的平均响度均等于从耳机的平均响度的情况，如果出现，则说明主耳机和从耳机所处的环境噪声相同，此时选择哪个耳机进行后续用户语音采集均可，例如默认此时选择主耳机进行采集(此时无需主耳机发送指令至从耳机进行切换，简化了控制逻辑)。
采用本实施例提供的均值比较方式，控制方式简单，处理时间少，提高了耳机语音采集的实时性。
当然，除了采用上述方式进行环境噪声确定，还可以对主耳机采集的噪声数据进行最小二乘法拟合得到频响曲线(横轴为频率，纵轴为响度)，对从耳机采集的噪声数据进行最小二乘法拟合得到频响曲线，通过分析两个频响曲线是否相交，以及相交时的交点以及在上的曲线的频率带占比，或者不相交时哪个曲线在上可以确定环境噪声，进而确定所处环境噪声较小的耳机。这里还可以采用其他方式进行噪声确定，本申请在此不作特别的限定。
请参照图5，图5为本发明提供的另一种语音采集控制方法的流程图，该方法应用于从耳机，包括：
S51：接收主耳机在检测到用户说话时启动主耳机的MIC进行噪声采集并发送的启动命令；
S52：根据启动指令控制从耳机的MIC进行噪声采集，以便主耳机根据采集到的主耳机的噪声数据和从耳机的噪声数据确定所处环境噪声较小的耳机，并控制所处环境噪声较小的耳机的MIC进行用户语音采集。
作为一种优选地实施例，启动命令包括噪声采集的时间起始点和时间间隔；
根据启动指令控制从耳机的MIC进行噪声采集，包括：
控制从耳机的MIC从噪声采集的时间起始点开始噪声采集，并每隔时间间隔将采集到的噪声数据同步至主耳机。
作为一种优选地实施例，在所处环境噪声较小的耳机为从耳机时，还包括：
在用户语音采集结束后，发送语音采集完毕指令至主耳机，以将耳机的语音采集权切回至主耳机的MIC。
本实施例为与主耳机配对的从耳机的语音采集控制方法实施例，对于本实施例提供的应用于从耳机的语音采集控制方法的介绍请参照上述实施例，本发明在此不再赘述。
请参照图6，图6为本发明提供的一种语音采集控制装置的结构示意图，应用于主耳机，该装置包括：
启动模块61，用于在检测到用户说话时，启动主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便从耳机根据启动指令控制自身的MIC进行噪声采集；
噪声处理模块62，用于根据采集到的主耳机的噪声数据和从耳机的噪声数据确定所处环境噪声较小的耳机；
控制模块63，用于控制所处环境噪声较小的耳机的MIC进行用户语音采集。
本实施例为主耳机的语音采集控制方法对应地装置实施例，对于本实施例提供的应用于主耳机的语音采集控制装置的介绍请参照上述实施例，本发明在此不再赘述。
请参照图7，图7为本发明提供的另一种语音采集控制装置的结构示意图，应用于从耳机，该装置包括：
接收模块71，用于接收主耳机在检测到用户说话时启动主耳机的MIC进行噪声采集并发送的启动命令；
控制模块72，用于根据启动指令控制从耳机的MIC进行噪声采集，以便主耳机根据采集到的主耳机的噪声数据和从耳机的噪声数据确定所处环境噪声较小的耳机，并控制所处环境噪声较小的耳机的MIC进行用户语音采集。
本实施例为与主耳机配对的从耳机的语音采集控制方法对应地装置实施例，对于本实施例提供的应用于从耳机的语音采集控制装置的介绍请参照上述实施例，本发明在此不再赘述。
本发明还提供了一种主耳机的结构示意图，该主耳机包括：
MIC；
存储器，用于存储计算机程序；
处理器，用于执行计算机程序时实现如上述应用于主耳机的语音采集控制方法的步骤。
本实施例为主耳机的语音采集控制方法对应地主耳机实施例，对于本实施例提供的主耳机的介绍请参照上述实施例，本发明在此不再赘述。
本发明还提供了一种从耳机，包括：
MIC；
存储器，用于存储计算机程序；
处理器，用于执行计算机程序时实现如上述应用于从耳机的语音采集控制方法的步骤。
本实施例为从耳机的语音采集控制方法对应地从耳机实施例，对于本实施例提供的从耳机的介绍请参照上述实施例，本发明在此不再赘述。
请参照图8，图8为本发明提供的一种TWS耳机的结构示意图，该TWS耳机包括：
MIC 81；
存储器83，用于存储计算机程序；
处理器82，用于执行计算机程序时实现如上述所述的语音采集控制方法的步骤。
具体地，TWS包括主耳机和从耳机。对于本实施例提供的TWS(True Wireless，真无线)耳机的介绍请参照上述实施例，本发明在此不再赘述。
需要说明的是，在本说明书中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本发明。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况下，在其他实施例中实现。因此，本发明将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (13)

  1. 一种语音采集控制方法，应用于主耳机，其特征在于，包括：
    在检测到用户说话时，启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的MIC进行噪声采集；
    根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机；
    控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
  2. 如权利要求1所述的语音采集控制方法，其特征在于，所述启动命令包括噪声采集的时间起始点和时间间隔；
    所述从耳机根据所述启动指令进行噪声采集，包括：
    所述从耳机控制自身的MIC从所述噪声采集的时间起始点开始噪声采集，并每隔所述时间间隔将采集到的噪声数据同步至所述主耳机。
  3. 如权利要求1所述的语音采集控制方法，其特征在于，所述启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机之后，所述根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机之前，还包括：
    根据采集到的所述主耳机的噪声数据进行关键词检测；
    若检测到关键词，则发送关键词检测成功的指令至所述从耳机，以控制所述主耳机和所述从耳机停止噪声检测，并进行后续步骤；
    若没有检测到关键词，则发送关键词检测失败的指令至所述从耳机，以控制所述主耳机和所述从耳机停止噪声检测，并停止进行后续步骤，结束本次语音采集。
  4. 如权利要求1所述的语音采集控制方法，其特征在于，在所述所处环境噪声较小的耳机为从耳机时，还包括：
    接收所述从耳机在用户语音采集结束后发送的语音采集完毕指令后，将耳机的语音采集权切回至所述主耳机的MIC。
  5. 如权利要求1所述的语音采集控制方法，其特征在于，所述控制所述所处环境噪声较小的耳机的MIC进行用户语音采集之后，还包括：
    将采集到的用户语音进行处理后发送至终端的语音助手。
  6. 如权利要求1-5任一项所述的语音采集控制方法，其特征在于，所述主耳机和所述从耳机均包括设置于耳机本体外部的外部MIC和设置于耳机本体内部的内部MIC；
    所述在检测到用户说话时，启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的MIC进行噪声采集，包括：
    在通过所述主耳机的内部MIC检测到用户说话时，启动所述主耳机上的外部MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的外部MIC进行噪声采集；
    所述控制所述所处环境噪声较小的耳机的MIC进行用户语音采集，包括：
    控制所述所处环境噪声较小的耳机的外部MIC进行用户语音采集。
  7. 如权利要求1-5任一项所述的语音采集控制方法，其特征在于，所述根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机，包括：
    获取相同时间段内的所述主耳机的噪声数据和所述从耳机的噪声数据；
    根据所述时间段内的所述主耳机的噪声数据得到所述主耳机在不同频段对应的平均响度以及根据所述时间段内的所述从耳机的噪声数据得到与所述主耳机在相同频段的所述从耳机的平均响度；
    分别对所述不同频段的主耳机的平均响度与从耳机的平均响度进行比较，确定平均响度较小对应的频段数多的耳机作为所处环境噪声较小的耳机。
  8. 一种语音采集控制方法，应用于从耳机，其特征在于，包括：
    接收主耳机在检测到用户说话时启动所述主耳机的MIC进行噪声采集并发送的启动命令；
    根据所述启动指令控制所述从耳机的MIC进行噪声采集，以便所述主耳机根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机，并控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
  9. 如权利要求8所述的语音采集控制方法，其特征在于，所述启动命令包括噪声采集的时间起始点和时间间隔；
    所述根据所述启动指令控制所述从耳机的MIC进行噪声采集，包括：
    控制所述从耳机的MIC从所述噪声采集的时间起始点开始噪声采集，并每隔所述时间间隔将采集到的噪声数据同步至所述主耳机。
  10. 如权利要求8所述的语音采集控制方法，其特征在于，在所述所处环境噪声较小的耳机为从耳机时，还包括：
    在用户语音采集结束后，发送语音采集完毕指令至所述主耳机，以将耳机的语音采集权切回至所述主耳机的MIC。
  11. 一种语音采集控制装置，应用于主耳机，其特征在于，包括：
    启动模块，用于在检测到用户说话时，启动所述主耳机的MIC进行噪声采集并发送启动命令至从耳机，以便所述从耳机根据所述启动指令控制自身的MIC进行噪声采集；
    噪声处理模块，用于根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机；
    控制模块，用于控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
  12. 一种语音采集控制装置，应用于从耳机，其特征在于，包括：
    接收模块，用于接收主耳机在检测到用户说话时启动所述主耳机的MIC进行噪声采集并发送的启动命令；
    控制模块，用于根据所述启动指令控制所述从耳机的MIC进行噪声采集，以便所述主耳机根据采集到的所述主耳机的噪声数据和所述从耳机的噪声数据确定所处环境噪声较小的耳机，并控制所述所处环境噪声较小的耳机的MIC进行用户语音采集。
  13. 一种TWS耳机，其特征在于，包括：
    MIC；
    存储器，用于存储计算机程序；
    处理器，用于执行所述计算机程序时实现如权利要求1至10任一项所述语音采集控制方法的步骤。
PCT/CN2019/109022 2019-06-28 2019-09-29 语音采集控制方法、装置及tws耳机 WO2020258570A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/608,322 US11937055B2 (en) 2019-06-28 2019-09-29 Voice acquisition control method and device, and TWS earphones

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910576125.1A CN110166879B (zh) 2019-06-28 2019-06-28 语音采集控制方法、装置及tws耳机
CN201910576125.1 2019-06-28

Publications (1)

Publication Number Publication Date
WO2020258570A1 true WO2020258570A1 (zh) 2020-12-30

Family

ID=67637145

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/109022 WO2020258570A1 (zh) 2019-06-28 2019-09-29 语音采集控制方法、装置及tws耳机

Country Status (3)

Country Link
US (1) US11937055B2 (zh)
CN (1) CN110166879B (zh)
WO (1) WO2020258570A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4262236A4 (en) * 2021-01-13 2024-05-22 Samsung Electronics Co Ltd ELECTRONIC DEVICE FOR PROCESSING AUDIO DATA, AND ASSOCIATED OPERATING METHOD

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110166879B (zh) * 2019-06-28 2020-11-13 歌尔科技有限公司 语音采集控制方法、装置及tws耳机
CN111246336B (zh) * 2020-02-27 2022-03-08 深迪半导体(绍兴)有限公司 耳机和电子设备
CN112256234B (zh) * 2020-11-06 2021-07-23 易普森智慧健康科技（深圳）有限公司 一种控制音量的方法、装置及计算机可读介质
CN112399301B (zh) * 2020-11-18 2023-03-17 维沃移动通信有限公司 耳机及降噪方法
CN112233697B (zh) * 2020-12-09 2021-04-13 北京云测信息技术有限公司 音频数据检测方法、装置及音频数据检测设备
WO2022151651A1 (zh) * 2021-01-13 2022-07-21 神盾股份有限公司 语音助理系统
CN113329290B (zh) * 2021-05-18 2023-04-07 RealMe重庆移动通信有限公司 耳机控制方法、装置、无线耳机及存储介质
CN114666445B (zh) * 2022-02-25 2023-09-01 华为技术有限公司 通话方法、装置、芯片和可读存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011134212A1 (zh) * 2010-04-29 2011-11-03 中兴通讯股份有限公司 终端和语音输入方法
CN108766431A (zh) * 2018-06-06 2018-11-06 广东小天才科技有限公司 一种基于语音识别的自动唤醒方法及电子设备
CN108882087A (zh) * 2018-06-12 2018-11-23 歌尔科技有限公司 一种智能语音检测方法、无线耳机、tws耳机及终端
CN109151211A (zh) * 2018-09-30 2019-01-04 Oppo广东移动通信有限公司 语音处理方法、装置以及电子设备
CN109413526A (zh) * 2018-09-30 2019-03-01 Oppo广东移动通信有限公司 耳机控制方法、装置、电子设备以及存储介质
CN110166879A (zh) * 2019-06-28 2019-08-23 歌尔科技有限公司 语音采集控制方法、装置及tws耳机

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2945696B1 (fr) 2009-05-14 2012-02-24 Parrot Procede de selection d'un microphone parmi deux microphones ou plus, pour un systeme de traitement de la parole tel qu'un dispositif telephonique "mains libres" operant dans un environnement bruite.
CN105469819A (zh) * 2014-08-20 2016-04-06 中兴通讯股份有限公司 麦克选择方法及装置
CN104270489A (zh) * 2014-09-10 2015-01-07 中兴通讯股份有限公司 一种从多个麦克风中确定主副麦克风的方法和系统
CN105611014A (zh) * 2014-11-21 2016-05-25 中兴通讯股份有限公司 移动终端通话语音降噪方法及装置
US11043228B2 (en) * 2015-05-12 2021-06-22 Nec Corporation Multi-microphone signal processing apparatus, method, and program for wind noise suppression
CN105847497A (zh) * 2016-03-28 2016-08-10 乐视控股（北京）有限公司 一种语音信号处理方法及装置
CN106303804B (zh) * 2016-07-28 2019-11-19 维沃移动通信有限公司 一种麦克风的控制方法及移动终端
CN106331359B (zh) * 2016-08-31 2017-09-12 广东欧珀移动通信有限公司 一种语音信号采集方法、装置及终端
CN206533526U (zh) * 2016-12-02 2017-09-29 歌尔科技有限公司 一种手环
CN107452375A (zh) * 2017-07-17 2017-12-08 湖南海翼电子商务股份有限公司 蓝牙耳机
US10706868B2 (en) * 2017-09-06 2020-07-07 Realwear, Inc. Multi-mode noise cancellation for voice detection
US10885907B2 (en) * 2018-02-14 2021-01-05 Cirrus Logic, Inc. Noise reduction system and method for audio device with multiple microphones
US10535362B2 (en) * 2018-03-01 2020-01-14 Apple Inc. Speech enhancement for an electronic device
CN108886647B (zh) * 2018-04-04 2020-08-25 万魔声学科技有限公司 耳机降噪方法及装置、主耳机、从耳机及耳机降噪系统
CN109151644B (zh) * 2018-09-21 2020-04-14 万魔声学科技有限公司 耳机的主从状态分配方法和装置
JP6807134B2 (ja) * 2018-12-28 2021-01-06 日本電気株式会社 音声入出力装置、補聴器、音声入出力方法および音声入出力プログラム
CN109712626B (zh) * 2019-03-04 2021-04-30 腾讯科技（深圳）有限公司 一种语音数据处理方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011134212A1 (zh) * 2010-04-29 2011-11-03 中兴通讯股份有限公司 终端和语音输入方法
CN108766431A (zh) * 2018-06-06 2018-11-06 广东小天才科技有限公司 一种基于语音识别的自动唤醒方法及电子设备
CN108882087A (zh) * 2018-06-12 2018-11-23 歌尔科技有限公司 一种智能语音检测方法、无线耳机、tws耳机及终端
CN109151211A (zh) * 2018-09-30 2019-01-04 Oppo广东移动通信有限公司 语音处理方法、装置以及电子设备
CN109413526A (zh) * 2018-09-30 2019-03-01 Oppo广东移动通信有限公司 耳机控制方法、装置、电子设备以及存储介质
CN110166879A (zh) * 2019-06-28 2019-08-23 歌尔科技有限公司 语音采集控制方法、装置及tws耳机

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4262236A4 (en) * 2021-01-13 2024-05-22 Samsung Electronics Co Ltd ELECTRONIC DEVICE FOR PROCESSING AUDIO DATA, AND ASSOCIATED OPERATING METHOD

Also Published As

Publication number Publication date
CN110166879B (zh) 2020-11-13
CN110166879A (zh) 2019-08-23
US11937055B2 (en) 2024-03-19
US20220167084A1 (en) 2022-05-26

Similar Documents

Publication Publication Date Title
WO2020258570A1 (zh) 语音采集控制方法、装置及tws耳机
US10466961B2 (en) Method for processing audio signal and related products
CN106910500B (zh) 对带麦克风阵列的设备进行语音控制的方法及设备
CN106782589B (zh) 移动终端及其语音输入方法和装置
JP4812941B2 (ja) 注目期間を有する音声入力装置
KR102098136B1 (ko) 응답을 제공하기 위한 디바이스 선택
CN110493678B (zh) 耳机的控制方法、装置、耳机和存储介质
US10485049B1 (en) Wireless device connection handover
JP4837917B2 (ja) 音声に基づく装置制御
US8260618B2 (en) Method and apparatus for remote control of devices through a wireless headset using voice activation
CN110678921A (zh) 来自智能扬声器的动态文本到语音响应
US20100235161A1 (en) Simultaneous interpretation system
WO2020019821A1 (zh) 麦克风堵孔检测方法及相关产品
CN107465970B (zh) 用于语音通信的设备
WO2021184549A1 (zh) 单耳耳机、智能电子设备、方法和计算机可读介质
CN110709931B (zh) 用于音频模式识别的系统和方法
CN109379490B (zh) 音频播放方法、装置、电子设备及计算机可读介质
CN109871238A (zh) 语音交互方法、装置和存储介质
JP2017138536A (ja) 音声処理装置
CN110517702B (zh) 信号生成的方法、基于人工智能的语音识别方法及装置
CN114333886A (zh) 音频处理方法、装置、电子设备及存储介质
US20200160858A1 (en) Wake-up word detection
CN111741391B (zh) 真无线耳机及其敲击实现操作控制的方法、装置和系统
CN115038002A (zh) 一种蓝牙音频的播放方法、存储介质以及真无线耳机
CN114120603A (zh) 语音控制方法、耳机和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19934847

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19934847

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19934847

Country of ref document: EP

Kind code of ref document: A1