A headphone noise reduction method and system based on bone voiceprint sensor

Abstract

The present invention discloses a method and system for reducing noise in headphones based on a bone voiceprint sensor. The first environmental sound data is obtained through a main call MIC module, and noise is judged according to the first environmental sound data. If the noise meets a preset condition, the second environmental sound data is obtained through an auxiliary MIC module, and the first environmental sound data and the second environmental sound data are subjected to noise fusion analysis to obtain environmental noise data, and an inverse noise reduction analysis is performed on the environmental noise data to obtain corresponding noise reduction audio data. User voice data is collected through a bone voiceprint sensor, and voice quality analysis is performed according to the user voice data. The user voice data is corrected in combination with the auxiliary MIC module to obtain second user voice data, and the noise reduction audio data and the second user voice data are transmitted in real time through a preset headphone device. The present invention can realize rapid noise reduction in multiple scenarios of headphones and improve user experience.

Description

Technical Field

[0001] The present invention relates to the field of headphone noise reduction, and more specifically, to a headphone noise reduction method and system based on a bone voiceprint sensor. Background Art

[0002] Bone voiceprint sensors can be widely used in TWS headphones, smart watches, bracelets, VR, AR, medical, military and other fields, with voice noise reduction, voice wake-up and boneprint ID functions.

[0003] However, the existing TWS earphones have very poor call performance in high-noise environments, such as subways, airports, and high-speed rail stations with large crowds, and cannot eliminate noise below 70dB. Therefore, how to use bone voiceprint sensors to achieve better noise reduction calls is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present invention overcomes the defects of the prior art and proposes a headphone noise reduction method and system based on a bone voiceprint sensor.

[0005] A first aspect of the present invention provides a headphone noise reduction method based on a bone voiceprint sensor, comprising:

[0006] Acquire the first environment sound data through the main call MIC module;

[0007] Perform noise judgment according to the first environmental sound data, and if the noise meets the preset conditions, obtain the second environmental sound data through the auxiliary MIC module;

[0008] Performing noise fusion analysis on the first ambient sound data and the second ambient sound data to obtain ambient noise data, and performing inverse noise reduction analysis on the ambient noise data to obtain corresponding noise reduction audio data;

[0009] The user voice data is collected through the bone voiceprint sensor, the voice quality analysis is performed according to the user voice data, and the user voice data is corrected in combination with the auxiliary MIC module to obtain the second user voice data;

[0010] The noise reduction audio data and the second user's voice data are transmitted in real time through a preset earphone device.

[0011] In this solution, the first environment sound data is obtained through the main call MIC module, which includes:

[0012] Construct a noise feature database;

[0013] Obtain historical monitoring sound data and historical monitoring scene information;

[0014] According to the historical monitoring scene information, one of the scene information is obtained and marked as the first scene;

[0015] According to the first scenario, all corresponding monitoring sound data are acquired from the historical monitoring sound data to obtain a plurality of sample sound data;

[0016] Analyze and extract audio features from a plurality of sample sound data to obtain a plurality of audio feature data;

[0017] Perform noise similarity analysis on multiple audio feature data to obtain noise feature data with the highest similarity;

[0018] Associating the noise feature data with the first scene category information to form scene noise associated feature data;

[0019] Based on the historical monitoring scene information, the scene noise-related feature data corresponding to other scene information is calculated and analyzed, and all scene noise-related feature data are imported into the noise feature database.

[0020] In this solution, the noise is judged according to the first environmental sound data. If the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically:

[0021] Extracting audio features according to the first environmental sound data to obtain first audio features;

[0022] Perform feature matching retrieval on the first audio feature from the noise feature database, and if the matching degree is lower than a preset value, obtain the second environmental sound data through the auxiliary MIC module;

[0023] If the matching degree is greater than a preset value, the corresponding matching noise feature is obtained;

[0024] An inverse noise reduction analysis is performed according to the matching noise characteristics, and noise-reduced frequency data is obtained.

[0025] In this solution, the noise is judged according to the first environmental sound data. If the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically:

[0026] Perform feature matching retrieval on the first audio feature from the noise feature database, and if the matching degree is less than a preset value, obtain the second environmental sound data through the auxiliary MIC module;

[0027] Extracting audio features from the second ambient sound data to obtain second audio features;

[0028] Superimpose and fuse the first audio feature with the second audio feature to obtain a superimposed audio feature;

[0029] Perform feature matching retrieval on the superimposed audio features from the noise feature database to obtain the matching noise feature with the highest matching degree and the matching degree;

[0030] If the matching degree is greater than or equal to a second preset value, an inverse noise reduction analysis is performed according to the matching noise characteristics to obtain noise reduction frequency data.

[0031] In this solution, the noise is judged according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, which also includes:

[0032] Perform feature matching retrieval on the superimposed audio features from the noise feature database to obtain the matching noise feature with the highest matching degree and the matching degree;

[0033] If the matching degree is less than a second preset value, the superimposed audio feature is associated with the current environment category and imported into the noise feature database.

[0034] In this solution, the user voice data is collected through the bone voiceprint sensor, voice quality analysis is performed based on the user voice data, and the user voice data is corrected in combination with the auxiliary MIC module to obtain the second user voice data, specifically:

[0035] Obtain user vibration signals through bone voiceprint sensors;

[0036] converting the vibration signal into a sound signal;

[0037] Performing a preset audio standard detection on the sound signal, and if it meets the preset standard, converting the sound signal into user voice data for call transmission;

[0038] If the preset standard is not met, the first user voice data is obtained through the main call MIC module, and the first user voice data is converted into audio to obtain first audio data;

[0039] Converting the sound signal into bone voiceprint audio data;

[0040] The first audio data and the bone voiceprint audio data are superimposed and converted into voice data to obtain the second user voice data.

[0041] In this solution, the real-time transmission of the noise reduction audio data and the second user's voice data through the preset earphone device also includes:

[0042] The vibration signal of the user during the call is obtained through the bone voiceprint sensor;

[0043] Extracting vibration features from the vibration signal to obtain bone voiceprint feature information of the user;

[0044] The bone voiceprint feature information is evenly divided into N segments of comparative feature information;

[0045] The user's current vibration signal is obtained in real time through the bone voiceprint sensor;

[0046] Extract vibration features from the vibration signal to obtain current bone voiceprint feature information;

[0047] Divide the current bone voiceprint feature information into N segments of user feature information on an even basis;

[0048] Perform feature comparison analysis on the user feature information and the comparison feature information one by one, and obtain N feature similarities;

[0049] If the similarities of the N features are all greater than the preset similarity, the current earphone device is set to the human voice speaking mode, and the active call MIC module is turned on to collect voice data, otherwise it is set to the normal mode.

[0050] The second aspect of the present invention further provides an earphone noise reduction system based on a bone voiceprint sensor, the system comprising: a memory, a processor, the memory comprising an earphone noise reduction program based on a bone voiceprint sensor, the earphone noise reduction program based on a bone voiceprint sensor when executed by the processor implements the following steps:

[0051] Acquire the first environment sound data through the main call MIC module;

[0052] Perform noise judgment according to the first environmental sound data, and if the noise meets the preset conditions, obtain the second environmental sound data through the auxiliary MIC module;

[0053] Performing noise fusion analysis on the first ambient sound data and the second ambient sound data to obtain ambient noise data, and performing inverse noise reduction analysis on the ambient noise data to obtain corresponding noise reduction audio data;

[0054] The user voice data is collected through the bone voiceprint sensor, the voice quality analysis is performed according to the user voice data, and the user voice data is corrected in combination with the auxiliary MIC module to obtain the second user voice data;

[0055] The noise reduction audio data and the second user's voice data are transmitted in real time through a preset earphone device.

[0056] In this solution, the first environment sound data is obtained through the main call MIC module, which includes:

[0057] Construct a noise feature database;

[0058] Obtain historical monitoring sound data and historical monitoring scene information;

[0059] According to the historical monitoring scene information, one of the scene information is obtained and marked as the first scene;

[0060] According to the first scenario, all corresponding monitoring sound data are acquired from the historical monitoring sound data to obtain a plurality of sample sound data;

[0061] Analyze and extract audio features from a plurality of sample sound data to obtain a plurality of audio feature data;

[0062] Perform noise similarity analysis on multiple audio feature data to obtain noise feature data with the highest similarity;

[0063] Associating the noise feature data with the first scene category information to form scene noise associated feature data;

[0064] Based on the historical monitoring scene information, the scene noise-related feature data corresponding to other scene information is calculated and analyzed, and all scene noise-related feature data are imported into the noise feature database.

[0065] In this solution, the noise is judged according to the first environmental sound data. If the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically:

[0066] Extracting audio features according to the first environmental sound data to obtain first audio features;

[0067] Perform feature matching retrieval on the first audio feature from the noise feature database, and if the matching degree is lower than a preset value, obtain the second environmental sound data through the auxiliary MIC module;

[0068] If the matching degree is greater than a preset value, the corresponding matching noise feature is obtained;

[0069] An inverse noise reduction analysis is performed according to the matching noise characteristics, and noise-reduced frequency data is obtained.

[0070] The present invention discloses a method and system for reducing noise in headphones based on a bone voiceprint sensor. The first environmental sound data is obtained through a main call MIC module, and noise is judged according to the first environmental sound data. If the noise meets a preset condition, the second environmental sound data is obtained through an auxiliary MIC module, and the first environmental sound data and the second environmental sound data are subjected to noise fusion analysis to obtain environmental noise data, and an inverse noise reduction analysis is performed on the environmental noise data to obtain corresponding noise reduction audio data. User voice data is collected through a bone voiceprint sensor, and voice quality analysis is performed according to the user voice data. The user voice data is corrected in combination with the auxiliary MIC module to obtain second user voice data, and the noise reduction audio data and the second user voice data are transmitted in real time through a preset headphone device. The present invention can realize rapid noise reduction in multiple scenarios of headphones and improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A flow chart of a headphone noise reduction method based on a bone voiceprint sensor of the present invention is shown;

[0072] Figure 2 The flowchart of building the noise characteristic database of the present invention is shown;

[0073] Figure 3 The flowchart of obtaining the noise reduction audio data of the present invention is shown;

[0074] Figure 4 A block diagram of an earphone noise reduction system based on a bone voiceprint sensor of the present invention is shown. DETAILED DESCRIPTION

[0075] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0076] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0077] Figure 1 A flow chart of a headphone noise reduction method based on a bone voiceprint sensor of the present invention is shown.

[0078] like Figure 1 As shown, the first aspect of the present invention provides a headphone noise reduction method based on a bone voiceprint sensor, comprising:

[0079] S102, obtaining first environmental sound data through the main call MIC module;

[0080] S104, performing noise determination according to the first ambient sound data, and if the noise meets a preset condition, obtaining second ambient sound data through an auxiliary MIC module;

[0081] S106, performing noise fusion analysis on the first ambient sound data and the second ambient sound data to obtain ambient noise data, and performing inverse noise reduction analysis on the ambient noise data to obtain corresponding noise reduction audio data;

[0082] S108, collecting user voice data through the bone voiceprint sensor, performing voice quality analysis based on the user voice data, and correcting the user voice data in combination with the auxiliary MIC module to obtain second user voice data;

[0083] S110: transmitting the noise reduction audio data and the second user's voice data in real time through a preset earphone device.

[0084] It should be noted that the preset earphone device includes a Bluetooth module, a main call MIC module, an auxiliary MIC module, an audio output module, a memory and a bone voiceprint sensor. The preset earphone device can perform Bluetooth connection and data transmission with a preset mobile terminal device and a preset computer terminal device.

[0085] Figure 2 The flowchart of building the noise characteristic database of the present invention is shown.

[0086] According to an embodiment of the present invention, the acquiring of the first environmental sound data through the main call MIC module includes:

[0087] S202, constructing a noise feature database;

[0088] S204, obtaining historical monitoring sound data and historical monitoring scene information;

[0089] S206, obtaining one type of scene information according to the historical monitoring scene information and marking it as the first scene;

[0090] S208, acquiring all corresponding monitoring sound data from the historical monitoring sound data according to the first scenario to obtain a plurality of sample sound data;

[0091] S210, analyzing and extracting audio features from a plurality of sample sound data to obtain a plurality of audio feature data;

[0092] S212, performing noise similarity analysis on the plurality of audio feature data to obtain noise feature data with the highest similarity;

[0093] S214, associating the noise feature data with the first scene category information to form scene noise associated feature data;

[0094] S216, calculating and analyzing scene noise-related feature data corresponding to other scene information based on historical monitoring scene information, and importing all scene noise-related feature data into a noise feature database.

[0095] It should be noted that the system of the present invention can be placed in a mobile terminal device or a computer terminal device, and the preset earphone device can connect and transmit data with the system via Bluetooth. The present invention can provide a data comparison basis for subsequent noise judgment and noise reduction by constructing a noise feature database, thereby realizing the rapid noise reduction function of the earphone and quickly judging the scene mode of the current user.

[0096] Figure 3 The flowchart of obtaining the noise-reduced audio data of the present invention is shown.

[0097] According to an embodiment of the present invention, the noise is judged according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically:

[0098] S302, extracting audio features according to the first environmental sound data to obtain a first audio feature;

[0099] S304, performing feature matching retrieval on the first audio feature from a noise feature database, and if the matching degree is lower than a preset value, obtaining second environmental sound data through an auxiliary MIC module;

[0100] S306, if the matching degree is greater than a preset value, obtaining a corresponding matching noise feature;

[0101] S308, performing inverse noise reduction analysis according to the matching noise characteristics, and obtaining noise-reduced frequency data.

[0102] It should be noted that the noise feature database is used to store noise feature data in different scenarios, including subways, airports, high-speed rail stations, indoors, outdoors, and scenes with high wind noise.

[0103] According to an embodiment of the present invention, the noise is judged according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically:

[0104] Perform feature matching retrieval on the first audio feature from the noise feature database, and if the matching degree is less than a preset value, obtain the second environmental sound data through the auxiliary MIC module;

[0105] Extracting audio features from the second ambient sound data to obtain second audio features;

[0106] Superimpose and fuse the first audio feature with the second audio feature to obtain a superimposed audio feature;

[0107] Perform feature matching retrieval on the superimposed audio features from the noise feature database to obtain the matching noise feature with the highest matching degree and the matching degree;

[0108] If the matching degree is greater than or equal to a second preset value, an inverse noise reduction analysis is performed according to the matching noise characteristics to obtain noise reduction frequency data.

[0109] It should be noted that when performing noise analysis through the first MIC module, it may be difficult to judge the noise characteristics due to the position of the MIC module and the error in the collected data. At this time, by adding an auxiliary MIC module for secondary environmental data collection and analysis, the accuracy of environmental noise judgment can be greatly improved, thereby improving the noise reduction capability of the headphones.

[0110] According to an embodiment of the present invention, the noise judgment is performed according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, and further includes:

[0111] Perform feature matching retrieval on the superimposed audio features from the noise feature database to obtain the matching noise feature with the highest matching degree and the matching degree;

[0112] If the matching degree is less than a second preset value, the superimposed audio feature is associated with the current environment category and imported into the noise feature database.

[0113] It should be noted that if the matching degree is less than the second preset value, it means that the current superimposed audio feature does not appear in the noise feature database. The present invention associates the superimposed audio feature with the current environment category and imports it into the noise feature database, and can store new noise environments and features in real time to enrich the noise data of the noise feature database.

[0114] According to an embodiment of the present invention, the user voice data is collected by the bone voiceprint sensor, voice quality analysis is performed according to the user voice data, and the user voice data is corrected in combination with the auxiliary MIC module to obtain the second user voice data, specifically:

[0115] Obtain user vibration signals through bone voiceprint sensors;

[0116] converting the vibration signal into a sound signal;

[0117] Performing a preset audio standard detection on the sound signal, and if it meets the preset standard, converting the sound signal into user voice data for call transmission;

[0118] If the preset standard is not met, the first user voice data is obtained through the main call MIC module, and the first user voice data is converted into audio to obtain first audio data;

[0119] Converting the sound signal into bone voiceprint audio data;

[0120] The first audio data and the bone voiceprint audio data are superimposed and converted into voice data to obtain the second user voice data.

[0121] According to an embodiment of the present invention, the real-time transmission of the noise reduction audio data and the second user voice data through a preset earphone device further includes:

[0122] The vibration signal of the user during the call is obtained through the bone voiceprint sensor;

[0123] Extracting vibration features from the vibration signal to obtain bone voiceprint feature information of the user;

[0124] The bone voiceprint feature information is evenly divided into N segments of comparative feature information;

[0125] The user's current vibration signal is obtained in real time through the bone voiceprint sensor;

[0126] Extract vibration features from the vibration signal to obtain current bone voiceprint feature information;

[0127] Divide the current bone voiceprint feature information into N segments of user feature information on an even basis;

[0128] Perform feature comparison analysis on the user feature information and the comparison feature information one by one, and obtain N feature similarities;

[0129] If the similarities of the N features are all greater than the preset similarity, the current earphone device is set to the human voice speaking mode, and the active call MIC module is turned on to collect voice data, otherwise it is set to the normal mode.

[0130] It should be noted that the normal mode is the normal monitoring mode of the headset, in which the headset does not input user voice data, and the human voice speaking mode is that the headset turns on the bone voiceprint sensor and the active call MIC module to collect user voice data. The present invention uses the bone voiceprint sensor to determine in real time and quickly whether the current user is speaking. If it is determined that the user is speaking, the corresponding module is turned on to collect voice data.

[0131] According to an embodiment of the present invention, the first ambient sound data and the second ambient sound data are subjected to noise fusion analysis to obtain ambient noise data, and the ambient noise data is subjected to inverse noise reduction analysis to obtain corresponding noise reduction audio data, further comprising:

[0132] Preset multiple scene modes and multiple scene mode information according to scene noise associated feature data in the noise feature database;

[0133] The scene mode information is consistent with the number of scene categories in the noise feature database;

[0134] Perform feature quality analysis and high-quality feature extraction on scene noise-related feature data to screen features;

[0135] The screening features are subjected to data compression and the data volume is compressed to within a preset data volume to obtain preloaded data.

[0136] It should be noted that the preloaded data has a relatively small amount of data and can be quickly loaded through the headphone device. The preset data amount is dynamically set by the system according to the data storage capacity of the headphone device.

[0137] According to an embodiment of the present invention, it also includes:

[0138] Monitor and record the user's environmental sound data within a preset number of days;

[0139] According to the user's environmental sound data, the scenario information corresponding to the user's daily noise data is analyzed, and the corresponding duration under different usage scenarios is calculated;

[0140] The scene information and duration are counted and summarized to obtain the user's multi-scene usage information;

[0141] Predict user scenarios based on user multi-scenario usage information to obtain the user's predicted scenario information within the current time;

[0142] Based on the predicted scene information, the headphone device is adjusted to the corresponding preset scene mode, and scene preloading data is obtained according to the corresponding scene for data preloading.

[0143] It should be noted that among users who use headphones frequently, the users often have certain life patterns, and the scenes in which they use headphones and the corresponding duration generally show periodic patterns. For example, a user regularly uses headphones when taking the subway after get off work. At this time, in the time period after get off work, the scene in which the user uses headphones is often fixed to the subway, and the duration is also relatively fixed. The present invention records the user's environmental sound data within a preset number of days and further predicts the user's usage scenario. It can set a suitable scene model and preload corresponding data in advance when the user uses the headphones, thereby improving the processing efficiency of headphone noise reduction and calls.

[0144] Figure 4 A block diagram of an earphone noise reduction system based on a bone voiceprint sensor of the present invention is shown.

[0145] The second aspect of the present invention further provides a headphone noise reduction system 4 based on a bone voiceprint sensor, the system comprising: a memory 41, a processor 42, the memory comprising a headphone noise reduction program based on a bone voiceprint sensor, the headphone noise reduction program based on a bone voiceprint sensor when executed by the processor implements the following steps:

[0146] Acquire the first environment sound data through the main call MIC module;

[0147] Perform noise judgment according to the first environmental sound data, and if the noise meets the preset conditions, obtain the second environmental sound data through the auxiliary MIC module;

[0148] Performing noise fusion analysis on the first ambient sound data and the second ambient sound data to obtain ambient noise data, and performing inverse noise reduction analysis on the ambient noise data to obtain corresponding noise reduction audio data;

[0149] The user voice data is collected through the bone voiceprint sensor, the voice quality analysis is performed according to the user voice data, and the user voice data is corrected in combination with the auxiliary MIC module to obtain the second user voice data;

[0150] The noise reduction audio data and the second user's voice data are transmitted in real time through a preset earphone device.

[0151] It should be noted that the preset earphone device includes a Bluetooth module, a main call MIC module, an auxiliary MIC module, an audio output module, a memory and a bone voiceprint sensor. The preset earphone device can perform Bluetooth connection and data transmission with a preset mobile terminal device and a preset computer terminal device.

[0152] According to an embodiment of the present invention, the acquiring of the first environmental sound data through the main call MIC module includes:

[0153] Construct a noise feature database;

[0154] Obtain historical monitoring sound data and historical monitoring scene information;

[0155] According to the historical monitoring scene information, one of the scene information is obtained and marked as the first scene;

[0156] According to the first scenario, all corresponding monitoring sound data are acquired from the historical monitoring sound data to obtain a plurality of sample sound data;

[0157] Analyze and extract audio features from a plurality of sample sound data to obtain a plurality of audio feature data;

[0158] Perform noise similarity analysis on multiple audio feature data to obtain noise feature data with the highest similarity;

[0159] Associating the noise feature data with the first scene category information to form scene noise associated feature data;

[0160] Based on the historical monitoring scene information, the scene noise-related feature data corresponding to other scene information is calculated and analyzed, and all scene noise-related feature data are imported into the noise feature database.

[0161] It should be noted that the system of the present invention can be placed in a mobile terminal device or a computer terminal device, and the preset earphone device can connect and transmit data with the system via Bluetooth. The present invention can provide a data comparison basis for subsequent noise judgment and noise reduction by constructing a noise feature database, thereby realizing the rapid noise reduction function of the earphone and quickly judging the scene mode of the current user.

[0162] According to an embodiment of the present invention, the noise is judged according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically:

[0163] Extracting audio features according to the first environmental sound data to obtain first audio features;

[0164] Perform feature matching retrieval on the first audio feature from the noise feature database, and if the matching degree is lower than a preset value, obtain the second environmental sound data through the auxiliary MIC module;

[0165] If the matching degree is greater than a preset value, the corresponding matching noise feature is obtained;

[0166] An inverse noise reduction analysis is performed according to the matching noise characteristics, and noise-reduced frequency data is obtained.

[0167] It should be noted that the noise feature database is used to store noise feature data in different scenarios, including subways, airports, high-speed rail stations, indoors, outdoors, and scenes with high wind noise.

[0168] According to an embodiment of the present invention, the noise is judged according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically:

[0169] Perform feature matching retrieval on the first audio feature from the noise feature database, and if the matching degree is less than a preset value, obtain the second environmental sound data through the auxiliary MIC module;

[0170] Extracting audio features from the second ambient sound data to obtain second audio features;

[0171] Superimpose and fuse the first audio feature with the second audio feature to obtain a superimposed audio feature;

[0172] Perform feature matching retrieval on the superimposed audio features from the noise feature database to obtain the matching noise feature with the highest matching degree and the matching degree;

[0173] If the matching degree is greater than or equal to a second preset value, an inverse noise reduction analysis is performed according to the matching noise characteristics to obtain noise reduction frequency data.

[0174] It should be noted that when performing noise analysis through the first MIC module, it may be difficult to judge the noise characteristics due to the position of the MIC module and the error in the collected data. At this time, by adding an auxiliary MIC module for secondary environmental data collection and analysis, the accuracy of environmental noise judgment can be greatly improved, thereby improving the noise reduction capability of the headphones.

[0175] According to an embodiment of the present invention, the noise judgment is performed according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, and further includes:

[0176] Perform feature matching retrieval on the superimposed audio features from the noise feature database to obtain the matching noise feature with the highest matching degree and the matching degree;

[0177] If the matching degree is less than a second preset value, the superimposed audio feature is associated with the current environment category and imported into the noise feature database.

[0178] It should be noted that if the matching degree is less than the second preset value, it means that the current superimposed audio feature does not appear in the noise feature database. The present invention associates the superimposed audio feature with the current environment category and imports it into the noise feature database, and can store new noise environments and features in real time to enrich the noise data of the noise feature database.

[0179] According to an embodiment of the present invention, the user voice data is collected by the bone voiceprint sensor, voice quality analysis is performed according to the user voice data, and the user voice data is corrected in combination with the auxiliary MIC module to obtain the second user voice data, specifically:

[0180] Obtain user vibration signals through bone voiceprint sensors;

[0181] converting the vibration signal into a sound signal;

[0182] Performing a preset audio standard detection on the sound signal, and if it meets the preset standard, converting the sound signal into user voice data for call transmission;

[0183] If the preset standard is not met, the first user voice data is obtained through the main call MIC module, and the first user voice data is converted into audio to obtain first audio data;

[0184] Converting the sound signal into bone voiceprint audio data;

[0185] The first audio data and the bone voiceprint audio data are superimposed and converted into voice data to obtain the second user voice data.

[0186] According to an embodiment of the present invention, the real-time transmission of the noise reduction audio data and the second user voice data through a preset earphone device further includes:

[0187] The vibration signal of the user during the call is obtained through the bone voiceprint sensor;

[0188] Extracting vibration features from the vibration signal to obtain bone voiceprint feature information of the user;

[0189] The bone voiceprint feature information is evenly divided into N segments of comparative feature information;

[0190] The user's current vibration signal is obtained in real time through the bone voiceprint sensor;

[0191] Extract vibration features from the vibration signal to obtain current bone voiceprint feature information;

[0192] Divide the current bone voiceprint feature information into N segments of user feature information on an even basis;

[0193] Perform feature comparison analysis on the user feature information and the comparison feature information one by one, and obtain N feature similarities;

[0194] If the similarities of the N features are all greater than the preset similarity, the current earphone device is set to the human voice speaking mode, and the active call MIC module is turned on to collect voice data, otherwise it is set to the normal mode.

[0195] It should be noted that the normal mode is the normal monitoring mode of the headset, in which the headset does not input user voice data, and the human voice speaking mode is that the headset turns on the bone voiceprint sensor and the active call MIC module to collect user voice data. The present invention uses the bone voiceprint sensor to determine in real time and quickly whether the current user is speaking. If it is determined that the user is speaking, the corresponding module is turned on to collect voice data.

[0196] According to an embodiment of the present invention, the first ambient sound data and the second ambient sound data are subjected to noise fusion analysis to obtain ambient noise data, and the ambient noise data is subjected to inverse noise reduction analysis to obtain corresponding noise reduction audio data, further comprising:

[0197] Preset multiple scene modes and multiple scene mode information according to scene noise associated feature data in the noise feature database;

[0198] The scene mode information is consistent with the number of scene categories in the noise feature database;

[0199] Perform feature quality analysis and high-quality feature extraction on scene noise-related feature data to screen features;

[0200] The screening features are subjected to data compression and the data volume is compressed to within a preset data volume to obtain preloaded data.

[0201] It should be noted that the preloaded data has a relatively small amount of data and can be quickly loaded through the headphone device. The preset data amount is dynamically set by the system according to the data storage capacity of the headphone device.

[0202] According to an embodiment of the present invention, it also includes:

[0203] Monitor and record the user's environmental sound data within a preset number of days;

[0204] According to the user's environmental sound data, the scenario information corresponding to the user's daily noise data is analyzed, and the corresponding duration under different usage scenarios is calculated;

[0205] The scene information and duration are counted and summarized to obtain the user's multi-scene usage information;

[0206] Predict user scenarios based on user multi-scenario usage information to obtain the user's predicted scenario information within the current time;

[0207] Based on the predicted scene information, the headphone device is adjusted to the corresponding preset scene mode, and scene preloading data is obtained according to the corresponding scene for data preloading.

[0208] It should be noted that among users who use headphones frequently, the users often have certain life patterns, and the scenes in which they use headphones and the corresponding duration generally show periodic patterns. For example, a user regularly uses headphones when taking the subway after get off work. At this time, in the time period after get off work, the scene in which the user uses headphones is often fixed to the subway, and the duration is also relatively fixed. The present invention records the user's environmental sound data within a preset number of days and further predicts the user's usage scenario. It can set a suitable scene model and preload corresponding data in advance when the user uses the headphones, thereby improving the processing efficiency of headphone noise reduction and calls.

[0209] The present invention discloses a method and system for reducing noise in headphones based on a bone voiceprint sensor. The first environmental sound data is obtained through a main call MIC module, and noise is judged according to the first environmental sound data. If the noise meets a preset condition, the second environmental sound data is obtained through an auxiliary MIC module, and the first environmental sound data and the second environmental sound data are subjected to noise fusion analysis to obtain environmental noise data, and an inverse noise reduction analysis is performed on the environmental noise data to obtain corresponding noise reduction audio data. User voice data is collected through a bone voiceprint sensor, and voice quality analysis is performed according to the user voice data. The user voice data is corrected in combination with the auxiliary MIC module to obtain second user voice data, and the noise reduction audio data and the second user voice data are transmitted in real time through a preset headphone device. The present invention can realize rapid noise reduction in multiple scenarios of headphones and improve user experience.

[0210] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0211] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0212] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0213] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0214] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0215] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A headphone noise reduction method based on bone voiceprint sensor, characterized in that: include: Acquire the first environment sound data through the main call MIC module; Perform noise judgment according to the first environmental sound data, and if the noise meets the preset conditions, obtain the second environmental sound data through the auxiliary MIC module; Performing noise fusion analysis on the first ambient sound data and the second ambient sound data to obtain ambient noise data, and performing inverse noise reduction analysis on the ambient noise data to obtain corresponding noise reduction audio data; The user voice data is collected through the bone voiceprint sensor, the voice quality analysis is performed according to the user voice data, and the user voice data is corrected in combination with the auxiliary MIC module to obtain the second user voice data; The noise reduction audio data and the second user's voice data are transmitted in real time through a preset earphone device; The method of acquiring the first environmental sound data through the main call MIC module includes: Construct a noise feature database; Obtain historical monitoring sound data and historical monitoring scene information; According to the historical monitoring scene information, one of the scene information is obtained and marked as the first scene; According to the first scenario, all corresponding monitoring sound data are acquired from the historical monitoring sound data to obtain a plurality of sample sound data; Analyze and extract audio features from a plurality of sample sound data to obtain a plurality of audio feature data; Perform noise similarity analysis on multiple audio feature data to obtain noise feature data with the highest similarity; Associating the noise feature data with the first scene category information to form scene noise associated feature data; Based on the historical monitoring scene information, calculate and analyze the scene noise correlation feature data corresponding to other scene information, and import all the scene noise correlation feature data into the noise feature database; The noise is judged according to the first environmental sound data, and if the noise meets the preset conditions, the second environmental sound data is obtained through the auxiliary MIC module, specifically: Extracting audio features according to the first environmental sound data to obtain first audio features; Performing feature matching retrieval on the first audio feature from a noise feature database, and if the obtained first matching degree is greater than a preset value, obtaining a corresponding matching noise feature; Perform inverse noise reduction analysis according to the matching noise characteristics and obtain noise reduction frequency data; If the first matching degree is lower than a preset value, obtaining second environmental sound data through the auxiliary MIC module; Extracting audio features from the second ambient sound data to obtain second audio features; Superimpose and fuse the first audio feature with the second audio feature to obtain a superimposed audio feature; Perform feature matching retrieval on the superimposed audio features from the noise feature database to obtain the matching noise feature with the highest matching degree and the second matching degree; If the second matching degree is greater than or equal to a second preset value, performing an inverse noise reduction analysis according to the matching noise characteristics, and obtaining noise reduction frequency data; If the second matching degree is less than a second preset value, associating the superimposed audio feature with the current environment category and importing the feature into a noise feature database; The method of collecting user voice data through the bone voiceprint sensor, performing voice quality analysis based on the user voice data, and correcting the user voice data in combination with the auxiliary MIC module to obtain the second user voice data is specifically as follows: Obtain user vibration signals through bone voiceprint sensors; converting the vibration signal into a sound signal; Performing a preset audio standard detection on the sound signal, and if it meets the preset standard, converting the sound signal into user voice data for call transmission; If the preset standard is not met, the first user voice data is obtained through the main call MIC module, and the first user voice data is converted into audio to obtain first audio data; Converting the sound signal into bone voiceprint audio data; The first audio data and the bone voiceprint audio data are superimposed and converted into voice data to obtain the second user voice data.

2. The headphone noise reduction method based on bone voiceprint sensor according to claim 1, characterized in that: The real-time transmission of the noise reduction audio data and the second user's voice data through the preset earphone device also includes: The vibration signal of the user during the call is obtained through the bone voiceprint sensor; Extracting vibration features from the vibration signal to obtain bone voiceprint feature information of the user; The bone voiceprint feature information is evenly divided into N segments of comparative feature information; The user's current vibration signal is obtained in real time through the bone voiceprint sensor; Extract vibration features from the vibration signal to obtain current bone voiceprint feature information; Divide the current bone voiceprint feature information into N segments of user feature information on an even basis; Perform feature comparison analysis on the user feature information and the comparison feature information one by one, and obtain N feature similarities; If the similarities of the N features are all greater than the preset similarity, the current earphone device is set to the human voice speaking mode, and the active call MIC module is turned on to collect voice data, otherwise it is set to the normal mode.

3. A headphone noise reduction system based on bone voiceprint sensor, characterized in that: The system comprises: a memory and a processor, wherein the memory comprises a headphone noise reduction program based on a bone voiceprint sensor, and when the headphone noise reduction program based on a bone voiceprint sensor is executed by the processor, the steps of implementing the headphone noise reduction method based on a bone voiceprint sensor as described in any one of claims 1 to 2 are implemented.

Images