Remote controller and remote controller voice recognition method

Abstract

The present invention belongs to the field of voice processing technology, and discloses a remote controller and a remote controller voice recognition method. The remote controller includes: a first microphone, a second microphone, a third microphone and a touch screen, wherein the third microphone is close to the touch screen, the second microphone is located at the edge of the touch screen, the third microphone is located below the touch screen, and the third microphone does not have a sound outlet. By using the above method, a third microphone without a sound outlet is added below the touch screen, so that the third microphone can accurately collect vibration noise data of the touch screen area, and recognize the voice data of the user corresponding to the remote controller according to the vibration noise data, thereby reducing the interference of noise on the user's voice data, and at the same time improving the recognition accuracy of the user's voice obtained by the remote controller and the accuracy of the remote controller's far-field sound pickup.

Description

Technical Field

[0001] The present invention relates to the field of voice processing technology, and in particular to a remote controller and a remote controller voice recognition method. Background Art

[0002] With the development of technology, existing TV remote controls usually use dual microphone arrays to achieve far-field sound pickup. Using this solution, the remote control has the following structure, with two control modes: touch screen and voice wake-up. However, under the existing solution, the remote control often fails to wake up or gives an incorrect answer after a command is issued to the remote control. The average voice wake-up recognition rate of the remote control is low, which greatly affects the user experience.

[0003] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of the present invention is to provide a remote control and a remote control voice recognition method, aiming to solve the technical problem of low accuracy of remote control voice recognition in the prior art.

[0005] To achieve the above objectives, the present invention provides a remote control, which includes a first microphone, a second microphone, a third microphone and a touch screen, wherein the second microphone is located at the edge of the touch screen, the third microphone is located below the touch screen, and the third microphone has no sound outlet.

[0006] Optionally, the lengths of the sound outlet holes of the first microphone and the second microphone are in the range of 6 to 9.5 mm.

[0007] In addition, to achieve the above object, the present invention further proposes a remote control voice recognition method, the remote control voice recognition method is applied to the remote control as described above, and the remote control voice recognition method includes:

[0008] Acquire first sound data from the first microphone, second sound data from the second microphone, and vibration noise data from the third microphone;

[0009] filtering the second sound data according to the vibration noise data to obtain third sound data;

[0010] Speech recognition is performed based on the third sound data and the first sound data to obtain a target recognition result.

[0011] Optionally, filtering the second sound data according to the vibration noise data to obtain third sound data includes:

[0012] Performing an audio comparison between the vibration noise data and the second sound data to determine the comparison audio data in the second sound data;

[0013] The second sound data is subjected to noise reduction filtering according to the compared audio data to obtain third sound data.

[0014] Optionally, performing an audio comparison between the vibration noise data and the second sound data to determine the compared audio data in the second sound data includes:

[0015] Extracting initial audio features of the second sound data to obtain an initial audio feature sequence;

[0016] Extracting target audio features of the vibration noise data to obtain a target audio feature sequence;

[0017] Comparing the initial audio feature sequence with the target audio feature sequence to obtain a comparison result;

[0018] The compared audio data in the second sound data is determined according to the comparison result.

[0019] Optionally, determining the compared audio data in the second sound data according to the comparison result includes:

[0020] Identify a comparison audio feature sequence in the second sound data that is identical to the target audio feature sequence;

[0021] Acquire audio data corresponding to the compared audio feature sequence;

[0022] The audio data is used as comparison audio data in the second sound data.

[0023] Optionally, filtering the second sound data according to the vibration noise data to obtain third sound data includes:

[0024] The second sound data is subtracted according to the vibration noise data to obtain third sound data.

[0025] Optionally, performing speech recognition according to the third sound data and the first sound data to obtain a target recognition result includes:

[0026] Processing the third sound data and the first sound data according to a preset signal enhancement model to obtain a target speech signal;

[0027] Perform keyword speech recognition on the target speech signal to obtain a target recognition result.

[0028] Optionally, the processing the third sound data and the first sound data according to a preset signal enhancement model to obtain a target speech signal includes:

[0029] determining a target sound position according to the third sound data and the first sound data;

[0030] Signal enhancement is performed on the third sound data and the first sound data according to the target sound position and a preset signal enhancement model to obtain a target speech signal.

[0031] Optionally, after performing keyword speech recognition on the target speech signal to obtain a target recognition result, the method further includes:

[0032] Searching for a target voice instruction corresponding to the target recognition result in a preset database;

[0033] The target voice command is sent to the cloud server, so that the cloud server feeds back corresponding response data according to the target voice command.

[0034] Optionally, before acquiring the first sound data from the first microphone, the second sound data from the second microphone, and the vibration noise data from the third microphone, the method further includes:

[0035] When an audio signal is detected, obtaining an operating mode of the first microphone;

[0036] If the working mode is a low power consumption working mode, the low power consumption working mode is turned off.

[0037] The remote control in the present invention includes a first microphone, a second microphone, a third microphone and a touch screen, wherein the third microphone is close to the touch screen, the second microphone is located at the edge of the touch screen, the third microphone is located below the touch screen, and the third microphone has no sound outlet. By adding a third microphone without a sound outlet below the touch screen in the above manner, the third microphone can accurately collect vibration noise data of the touch screen area, and recognize the voice data of the user corresponding to the remote control according to the vibration noise data, thereby reducing the interference of noise on the user's voice data, and at the same time improving the recognition accuracy of the user's voice obtained by the remote control and the accuracy of the remote control's far-field sound pickup. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a structural schematic diagram of the remote controller of the present invention;

[0039] Figure 2 A schematic diagram of a flow chart of a first embodiment of a remote controller voice recognition method according to the present invention;

[0040] Figure 3FIG. 4 is a flow chart of a second embodiment of a remote controller voice recognition method according to the present invention.

[0041] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0042] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0043] It should be noted that if the embodiment of the present invention involves directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0044] The present invention provides a remote controller, referring to Figure 1 , the remote controller 1 comprises:

[0045] A first microphone 10, a second microphone 20, a third microphone 30 and a touch screen 40, wherein the second microphone 20 is located at the edge of the touch screen 40, the third microphone 30 is located below the touch screen 40, and the third microphone 30 has no sound outlet hole;

[0046] It should be noted that the first microphone 10 and the second microphone have holes opened on the outer shell of the remote control 1, and the first microphone 10 is located outside the touch screen 40 on the remote control 1, and is spaced apart from the second microphone and the second microphone 20. The second microphone is located at the edge of the touch screen 40 on the remote control 1, and the specific installation position can be set according to actual applications. The third microphone 30 does not have a hole opened on the outer shell of the remote control 1, but is closely attached to the bottom of the touch screen 40, and the touch screen 40 is located on the outer shell of the remote control 1.

[0047] It is understandable that, under normal circumstances, the shorter the length of the microphone's sound outlet, the lower the energy loss when the sound wave reaches the microphone. Usually, the length of the microphone's sound outlet is 9.5mm. If only the length of the sound outlet is changed, the required effect cannot be met; at the same time, when the user sends a command to the remote control, the remote control will also capture the audio data of the terminal device corresponding to the remote control. If an algorithm is used to filter out the audio data of the terminal device corresponding to the remote control when recognizing the user's command, the accuracy of voice recognition will still not be improved; again, in the existing solution, the first microphone is a zero-power digital microphone, and the second microphone is an ordinary digital microphone. Therefore, when the surrounding environment is lower than 45dB, the first microphone will automatically enter zero-power mode, and the first microphone requires a sound greater than 45dB to wake up and start working, or when waking up the first microphone, there will be a system delay, which affects the accuracy of voice recognition. When the low-power mode of the first microphone is turned off, the accuracy of voice recognition is not significantly improved, so this solution is designed.

[0048] In a specific implementation, the first microphone 10 in this embodiment is used to Get The second microphone 20 is used to obtain the second sound data finally obtained by the human voice when the user issues a voice command to the remote control 1 and the vibration when the user touches the touch screen 40 area. Since the third microphone 30 has no sound hole and is located below the touch screen 40, the third microphone 30 is used to obtain the vibration noise data when the user touches the touch screen 40 area. The touch screen 40 is used for the user to issue key commands to the remote control 1.

[0049] It should be understood that the smaller the length of the sound outlet of the microphone, the lower the energy loss of external sound waves when reaching the microphone. Therefore, in the remote control 1, the length of the sound outlet of the first microphone 10 and the second microphone 20 is in the range of 6 to 9.5 mm. However, in order to achieve a more accurate voice recognition rate, in this embodiment, the length of the sound outlet of the first microphone 10 and the second microphone 20 are both set to 6 mm.

[0050] The remote control in this embodiment includes a first microphone, a second microphone, a third microphone and a touch screen, wherein the third microphone is close to the touch screen, the second microphone is located at the edge of the touch screen, the third microphone is located below the touch screen, and the third microphone has no sound outlet. By adding a third microphone without a sound outlet below the touch screen in the above manner, the third microphone can accurately collect vibration noise data of the touch screen area, and recognize the voice data of the user corresponding to the remote control according to the vibration noise data, thereby reducing the interference of noise on the user's voice data, and at the same time improving the recognition accuracy of the user's voice obtained by the remote control and the accuracy of the remote control's far-field sound pickup.

[0051] In addition, an embodiment of the present invention provides a remote control voice recognition method, referring to Figure 2 , Figure 2 The figure is a flow chart of a first embodiment of a remote controller voice recognition method according to the present invention.

[0052] In this embodiment, the remote control voice recognition method includes the following steps:

[0053] Step S10: acquiring first sound data from the first microphone, second sound data from the second microphone, and vibration noise data from the third microphone.

[0054] It should be noted that the executor of this embodiment is a remote control. The controller of the remote control obtains the audio data corresponding to the first microphone, the second microphone and the third microphone, and performs noise reduction filtering on the second sound data of the second microphone based on the vibration noise data of the third microphone, thereby performing speech recognition based on the third sound data after noise reduction filtering and the first sound data obtained by the first microphone to obtain a target recognition result.

[0055] It is understandable that the first microphone acquires first sound data, which includes but is not limited to the human voice when the user issues a voice command to the remote control, and the second microphone acquires second sound data, which includes but is not limited to the human voice part when the user issues a voice command to the remote control and the vibration part when the user touches the touch screen area. The third microphone acquires vibration noise data, and since the third microphone is located below the touch screen and has no sound hole, the vibration noise data only includes the vibration noise when the user touches the touch screen area.

[0056] In a specific implementation, when audio data is received, first sound data captured by the first microphone, second sound data captured by the second microphone, and vibration noise data captured by the third microphone are obtained.

[0057] Step S20: filtering the second sound data according to the vibration noise data to obtain third sound data.

[0058] It should be noted that since the second sound data includes but is not limited to the vocal part when the user issues a voice command to the remote control and the vibration part when the user touches the touch screen area, it is necessary to perform noise reduction filtering on the second sound data based on the vibration noise data of the third microphone to obtain the third sound data that does not include the vibration part when the user touches the touch screen area.

[0059] Step S30: performing speech recognition according to the third sound data and the first sound data to obtain a target recognition result.

[0060] It should be noted that after obtaining the third sound data of the second microphone and the first sound data of the first microphone, the two audio data need to be processed, and speech recognition is performed based on the processed audio data to obtain the corresponding recognition result. For example, the current first sound data is audio A, and the third sound data is audio B. After processing audio A and audio B, audio C is obtained. Speech recognition is performed on audio C, and the target recognition result is "I want to turn on the TV".

[0061] It can be understood that in order to obtain accurate recognition results, further, the speech recognition is performed based on the third sound data and the first sound data to obtain the target recognition result, including: processing the third sound data and the first sound data according to a preset signal enhancement model to obtain a target speech signal; performing keyword speech recognition on the target speech signal to obtain a target recognition result.

[0062] It should be noted that the preset signal enhancement model is a model obtained by training sample audio data through a beamforming algorithm, which can merge and process multi-channel audio data, suppress interference signals in non-target directions, and enhance sound signals in the target direction.

[0063] It can be understood that after obtaining the first sound data of the first microphone and the third sound data of the second microphone, the first sound data and the third sound data are audio-merged through a preset signal enhancement model, and the sound signal in the target direction is enhanced to finally obtain a processed target voice signal.

[0064] In a specific implementation, after obtaining the target voice signal, the controller of the remote control needs to perform keyword voice recognition on the target voice signal to obtain the corresponding target recognition result. For example, the target voice signal may include repeated text when the user issues a command or keywords that do not exist in the preset text. Therefore, the target voice signal is text-segmented, and the segmented text is compared with the preset text to obtain the target text that exists in the preset text, and the target text is spliced ​​to obtain the target recognition result. For example, if the target voice signal is "I, I, um, want to, turn on, um, um, TV", the target recognition result is "I want to turn on the TV".

[0065] It can be understood that in order to obtain an accurate target speech signal, the third sound data and the first sound data are further processed according to a preset signal enhancement model to obtain a target speech signal, including: determining the target sound position according to the third sound data and the first sound data; performing signal enhancement on the third sound data and the first sound data according to the target sound position and a preset signal enhancement model to obtain a target speech signal.

[0066] It should be noted that since the distribution positions of the first microphone and the second microphone are different, the target sound position of the user can be determined based on the first sound data and the third sound data. After obtaining the target sound position, the third sound data and the first sound data can be merged according to the target sound position and the directionality of the array system formed by the preset signal enhancement model, and spatial filtering is performed at the same time to suppress spatial interference and environmental noise, improve the signal-to-noise ratio, and finally obtain the target voice signal after signal enhancement.

[0067] It can be understood that in order to improve the user experience, after obtaining the target recognition result, a corresponding response can be made according to the target recognition result. Furthermore, after performing keyword voice recognition on the target voice signal and obtaining the target recognition result, it also includes: searching for the target voice instruction corresponding to the target recognition result in a preset database; sending the target voice instruction to the cloud server, so that the cloud server feeds back corresponding response data according to the target voice instruction.

[0068] In a specific implementation, a mapping relationship between keywords and voice commands is stored in a preset database. Therefore, after obtaining the target recognition result, each keyword in the target recognition result is searched in the preset database to obtain the corresponding target voice command. For example, if the current target recognition result is "I want to turn on the TV", the command search is performed on each keyword "I want" and "turn on the TV", so that the corresponding target voice command is "start the TV".

[0069] It should be noted that after receiving the target voice command, the target voice command is sent to the cloud server corresponding to the remote control, so that the cloud server feeds back corresponding response data according to the target voice command. For example, if the current target voice command is "start the TV", the cloud server starts the TV.

[0070] It can be understood that in order to improve the responsiveness and wake-up rate of the remote control system, further, before obtaining the first sound data of the first microphone, the second sound data of the second microphone and the vibration noise data of the third microphone, it also includes: when an audio signal is detected, obtaining the working mode of the first microphone; if the working mode is a low power consumption working mode, turning off the low power consumption working mode.

[0071] In a specific implementation, since the first microphone is usually in a low power consumption mode, in the low power consumption mode, when the surrounding environment audio is lower than a preset decibel value, the first microphone is not started, and when the surrounding environment audio is not lower than the preset decibel value, the first microphone starts working. If the decibel of the user's voice is lower than the preset decibel value, the microphone may not be woken up. Therefore, when the remote control detects an audio signal, the working mode of the first microphone is obtained. If the working mode is a low power consumption mode, the low power consumption mode is turned off.

[0072] This embodiment obtains the first sound data of the first microphone, the second sound data of the second microphone, and the vibration noise data of the third microphone; filters the second sound data according to the vibration noise data to obtain the third sound data; performs voice recognition according to the third sound data and the first sound data to obtain the target recognition result. By adding a third microphone, the second sound data of the second microphone is filtered using the vibration noise data of the third microphone to obtain the third sound data after noise reduction, and voice recognition is performed based on the third sound data and the first sound data, thereby reducing the interference of noise on the user's audio, and at the same time improving the recognition accuracy of the user's voice obtained by the remote control and the accuracy of the remote control's far-field sound pickup.

[0073] refer to Figure 3 , Figure 3 The figure is a flow chart of a second embodiment of a remote controller voice recognition method according to the present invention.

[0074] Based on the first embodiment, step S20 in the remote controller voice recognition method of this embodiment includes:

[0075] Step S21: performing an audio comparison between the vibration noise data and the second sound data to determine the compared audio data in the second sound data.

[0076] It should be noted that since the second sound data includes but is not limited to the vocal part when the user issues a voice command to the remote control and the vibration part when the user touches the touch screen area, it is necessary to compare the vibration noise data and the second sound data to obtain the comparison audio data in the second sound data that is the same as the vibration noise data, and the comparison audio data is the vibration part when the user touches the touch screen area that is the same as the audio of the vibration noise data.

[0077] It can be understood that in order to obtain accurate comparison audio data, further, the vibration noise data and the second sound data are subjected to audio comparison to determine the comparison audio data in the second sound data, including: extracting the initial audio features of the second sound data to obtain an initial audio feature sequence; extracting the target audio features of the vibration noise data to obtain a target audio feature sequence; comparing the initial audio feature sequence with the target audio feature sequence to obtain a comparison result; and determining the comparison audio data in the second sound data according to the comparison result.

[0078] In a specific implementation, different audio features corresponding to the audio data, such as the short-time energy of the audio, the pitch and frequency of the time-frequency, are extracted from the initial audio features of the second sound data to obtain an initial audio feature sequence, and the target audio features of the vibration noise data are extracted to obtain a target audio feature sequence.

[0079] It should be noted that after obtaining the target audio feature sequence and the initial audio feature sequence, a sequence comparison is performed to obtain a comparison result after the sequence comparison, and the comparison audio data in the second sound data can be determined according to the comparison result.

[0080] It can be understood that in order to determine the comparison audio data based on the comparison result, further, determining the comparison audio data in the second sound data based on the comparison result includes: identifying the comparison audio feature sequence in the second sound data that is the same as the target audio feature sequence; obtaining the audio data corresponding to the comparison audio feature sequence; and using the audio data as the comparison audio data in the second sound data.

[0081] In a specific implementation, a compared audio feature sequence that is identical to the initial audio feature sequence and the target audio feature sequence is obtained according to the comparison result, and the audio data corresponding to the compared audio feature sequence in the second sound data is the compared audio data.

[0082] Step S22: performing noise reduction filtering on the second sound data according to the compared audio data to obtain third sound data.

[0083] It should be noted that after obtaining the comparison audio data, the comparison audio data in the second sound data is filtered and deleted, thereby completing the noise reduction process on the second sound data and finally obtaining the third sound data.

[0084] It can be understood that, in order to obtain the third sound data after noise reduction, further, step S20 includes: subtracting the second sound data according to the vibration noise data to obtain the third sound data.

[0085] In a specific implementation, after obtaining the second sound data acquired by the second microphone and the vibration noise data acquired by the third microphone, an algorithm is used to filter out the vibration noise data in the second microphone and subtract them to obtain the third sound data that does not include the vibration part when the user touches the touch screen area.

[0086] In this embodiment, the vibration noise data and the second sound data are compared by audio, and the compared audio data in the second sound data are determined; the second sound data is subjected to noise reduction filtering according to the compared audio data to obtain the third sound data. By comparing the vibration noise data with the second sound data by audio, and performing noise reduction filtering on the second sound data according to the compared audio data, the accuracy of noise reduction filtering is improved, thereby obtaining the third sound data without noise.

[0087] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0088] In addition, for technical details not fully described in this embodiment, reference can be made to the remote control voice recognition method provided in any embodiment of the present invention, and will not be repeated here.

[0089] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0090] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0091] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0092] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A remote control voice recognition method, characterized in that: The remote control voice recognition method is applied to a remote control, the remote control comprises a first microphone, a second microphone, a third microphone and a touch screen, the first microphone is located outside the touch screen, the second microphone is located at the edge of the touch screen, the third microphone is located below the touch screen, and the third microphone has no sound outlet hole; The remote controller voice recognition method comprises: Acquire first sound data from a first microphone, second sound data from a second microphone, and vibration noise data from a third microphone, wherein the second sound data includes a human voice when a user issues a voice command to a remote controller and vibration when a user touches a touch screen area; filtering the second sound data according to the vibration noise data to obtain third sound data; Performing speech recognition according to the third sound data and the first sound data to obtain a target recognition result; The performing speech recognition according to the third sound data and the first sound data to obtain a target recognition result includes: The third sound data and the first sound data are processed according to a preset signal enhancement model to obtain a target speech signal, wherein the preset signal enhancement model is a model that is obtained by training sample audio data through a beamforming algorithm and is capable of merging multiple audio data, suppressing interference signals in non-target directions, and enhancing sound signals in the target direction; Perform keyword speech recognition on the target speech signal to obtain a target recognition result.

2. The remote controller voice recognition method according to claim 1, characterized in that: The lengths of the sound outlet holes of the first microphone and the second microphone are in the range of 6 to 9.5 mm.

3. The remote controller voice recognition method according to claim 1, characterized in that: The filtering the second sound data according to the vibration noise data to obtain the third sound data includes: Performing an audio comparison between the vibration noise data and the second sound data to determine the comparison audio data in the second sound data; The second sound data is subjected to noise reduction filtering according to the compared audio data to obtain third sound data.

4. The remote controller voice recognition method according to claim 3, characterized in that: The step of performing an audio comparison between the vibration noise data and the second sound data to determine the compared audio data in the second sound data includes: Extracting initial audio features of the second sound data to obtain an initial audio feature sequence; Extracting target audio features of the vibration noise data to obtain a target audio feature sequence; Comparing the initial audio feature sequence with the target audio feature sequence to obtain a comparison result; The compared audio data in the second sound data is determined according to the comparison result.

5. The remote controller voice recognition method according to claim 4, characterized in that: The determining the compared audio data in the second sound data according to the comparison result includes: Identify a comparison audio feature sequence in the second sound data that is identical to the target audio feature sequence; Acquire audio data corresponding to the compared audio feature sequence; The audio data is used as comparison audio data in the second sound data.

6. The remote controller voice recognition method according to claim 1, characterized in that: The filtering the second sound data according to the vibration noise data to obtain the third sound data includes: The second sound data is subtracted according to the vibration noise data to obtain third sound data.

7. The remote controller voice recognition method according to claim 6, characterized in that: The processing of the third sound data and the first sound data according to a preset signal enhancement model to obtain a target speech signal includes: determining a target sound position according to the third sound data and the first sound data; Signal enhancement is performed on the third sound data and the first sound data according to the target sound position and a preset signal enhancement model to obtain a target speech signal.

8. The remote controller voice recognition method according to claim 6, characterized in that: After performing keyword speech recognition on the target speech signal to obtain a target recognition result, the method further includes: Searching for a target voice instruction corresponding to the target recognition result in a preset database; The target voice command is sent to the cloud server, so that the cloud server feeds back corresponding response data according to the target voice command.

9. The remote controller voice recognition method according to any one of claims 1 to 8, characterized in that: Before acquiring the first sound data of the first microphone, the second sound data of the second microphone, and the vibration noise data of the third microphone, the method further includes: When an audio signal is detected, obtaining an operating mode of the first microphone; If the working mode is a low power consumption working mode, the low power consumption working mode is turned off.

Images