CN109346055A - Active denoising method, device, earphone and computer storage medium - Google Patents
Active denoising method, device, earphone and computer storage medium Download PDFInfo
- Publication number
- CN109346055A CN109346055A CN201811407883.2A CN201811407883A CN109346055A CN 109346055 A CN109346055 A CN 109346055A CN 201811407883 A CN201811407883 A CN 201811407883A CN 109346055 A CN109346055 A CN 109346055A
- Authority
- CN
- China
- Prior art keywords
- scene
- audio data
- noise reduction
- audio
- active
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 230000009467 reduction Effects 0.000 claims abstract description 85
- 238000013145 classification model Methods 0.000 claims abstract description 43
- 238000000605 extraction Methods 0.000 claims description 24
- 238000012545 processing Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 9
- 230000015654 memory Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 10
- 239000000284 extract Substances 0.000 description 6
- 238000009432 framing Methods 0.000 description 6
- 230000002441 reversible effect Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000001073 episodic memory Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000005036 nerve Anatomy 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000001755 vocal effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
- G10K11/1783—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase handling or detecting of non-standard events or conditions, e.g. changing operating modes under specific operating conditions
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
- G10K11/1785—Methods, e.g. algorithms; Devices
- G10K11/17853—Methods, e.g. algorithms; Devices of the filter
- G10K11/17854—Methods, e.g. algorithms; Devices of the filter the filter being an adaptive filter
-
- 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
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K2210/00—Details of active noise control [ANC] covered by G10K11/178 but not provided for in any of its subgroups
- G10K2210/30—Means
- G10K2210/301—Computational
- G10K2210/3035—Models, e.g. of the acoustic system
- G10K2210/30351—Identification of the environment for applying appropriate model characteristics
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K2210/00—Details of active noise control [ANC] covered by G10K11/178 but not provided for in any of its subgroups
- G10K2210/30—Means
- G10K2210/301—Computational
- G10K2210/3048—Pretraining, e.g. to identify transfer functions
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K2210/00—Details of active noise control [ANC] covered by G10K11/178 but not provided for in any of its subgroups
- G10K2210/30—Means
- G10K2210/301—Computational
- G10K2210/3055—Transfer function of the acoustic system
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R2201/00—Details of transducers, loudspeakers or microphones covered by H04R1/00 but not provided for in any of its subgroups
- H04R2201/10—Details of earpieces, attachments therefor, earphones or monophonic headphones covered by H04R1/10 but not provided for in any of its subgroups
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
Abstract
The present invention provides a kind of active denoising method, device, earphone and computer storage mediums, which comprises obtains audio data；Scene type corresponding to the audio data is judged using trained scene classification model, and exports the classification results of scene type；Active noise reduction mode is selected according to the classification results.The present invention selects noise reduction mode according to scene type, so as to avoid noise reduction bring insecurity, and avoids the triviality of manual switching noise reduction mode.
Description
Technical field
The present invention relates to audio signal processing technique field, relate more specifically to a kind of active denoising method, device, earphone and meter
Calculation machine storage medium.
Background technique
Ambient noise is the key factor for influencing earphone wearer sound quality sense and receiving.Earphone is worn in high-noise environment
It listens audio or carries out voice communication, will lead to user listens sound quality to decline, more serious to make user's impaired hearing.Cause
This, concern of the anti-acoustic capability of earphone by headset designs person and user.Currently, there are two types of modes for earphone noise reduction: passive noise reduction
And active noise reduction.Passive noise reduction realized generally by the structure of earphone, design of material earphone and extraneous physical signal every
From such as In-Ear design.And active noise reduction is that the method for taking analog or digital signal to handle offsets outside noise, one
As be using signal reversely be superimposed eliminate principle, with Mike reception ambient noise, calculate the reverse signal of ambient noise, and utilize
The reverse signal of loudspeaker launch environment noise, to offset outside noise.
Existing active noise reduction techniques have the following problems:
1, whether useful to user active noise reduction algorithm does not judge certain section of outside noise, can only receive to all
Sound carries out noise reduction, therefore environment (such as when going across the road) may be because the sound that can't hear road using noise cancelling headphone outdoors
And accident occurs；
2, existing active denoising method uses identical noise reduction mode in different environment mostly, and user exists more
The puzzlement for wanting to hear sound can not be heard under kind complex scene, and the earphone of manual switching noise reduction mode uses cumbersome, user's body
It tests bad.
Summary of the invention
Existing noise cancelling headphone carries out noise reduction for all noises, and can bring to user can not hear the tired of useful sound
It disturbs, and manual switching noise reduction mode is using cumbersome.
Based on the above-mentioned problems in the prior art, one aspect of the present invention provides a kind of active denoising method, the side
Method includes:
Obtain audio data；
Scene type corresponding to the audio data is judged using trained scene classification model, and exports scene class
Other classification results；And
Active noise reduction mode is selected according to the classification results.
In one embodiment, the training method of the scene classification model includes:
Feature extraction is carried out to the original audio data for having marked scene type in audio database, obtains original audio
Data characteristics and its corresponding type feature；
The original audio data feature and its corresponding type feature are input in deep neural network and are trained
To obtain initial scene classification model.
In one embodiment, the training method of the scene classification model further include:
After obtaining the initial scene classification model, audio data is expanded in acquisition, and collects the expansion audio
The associated audio data of data；
The expansion audio data and the associated audio data for expanding audio data are input to the depth nerve
It is trained in network, so that study expansion is carried out to the initial scene classification model, to obtain the trained field
Scape disaggregated model.
In one embodiment, the original audio data and/or the expansion audio data include: road surface sound and/
Or the sound that high frequency time repeats.
In one embodiment, the scene type includes complex scene and non-complex scene, described according to the classification
As a result selection noise reduction mode includes:
When the scene type is non-complex scene, common noise reduction mode is selected, to carry out noise reduction to whole noises；
When the scene type is complex scene, the noise reduction mode under complex scene is selected.
In one embodiment, the noise reduction mode under the complex scene includes:
Distinguish the useful audio and useless audio in the audio data；
Active noise reduction processing is carried out to the useless audio, to the useful audio without active noise reduction processing.
In one embodiment, described to judge field corresponding to the audio data using trained scene classification model
Scape classification, and the classification results for exporting scene type include:
Feature extraction is carried out to the audio data, to obtain audio frequency characteristics；
The audio frequency characteristics are inputted into the trained scene classification model, to obtain the classification knot of the scene type
Fruit.
According to a further aspect of the invention, a kind of active noise reducing device is provided, the active noise reducing device includes:
Audio obtains module, for obtaining audio data；
Scene classification module, for judging scene corresponding to the audio data using trained scene classification model
Classification, and export the classification results of scene type；And
Mode selection module, for selecting active noise reduction mode according to the classification results.
Another aspect according to the present invention, provides a kind of earphone, and the earphone includes storage device and processor, described to deposit
The computer program run by the processor is stored on storage device, the computer program by the processor when being run
Execute active denoising method of the invention.
In accordance with a further aspect of the present invention, a kind of computer storage medium is provided, computer program is stored thereon with, institute
State the step of active denoising method according to the present invention and each example the method are realized when program is executed by processor.
Active denoising method, device, earphone and computer storage medium of the invention selects noise reduction mould according to scene type
Formula so as to avoid noise reduction bring insecurity, and avoids the triviality of manual switching noise reduction mode.
Detailed description of the invention
The embodiment of the present invention is described in more detail in conjunction with the accompanying drawings, the above and other purposes of the present invention,
Feature and advantage will be apparent.Attached drawing is used to provide to further understand the embodiment of the present invention, and constitutes explanation
A part of book, is used to explain the present invention together with the embodiment of the present invention, is not construed as limiting the invention.In the accompanying drawings,
Identical reference label typically represents same parts or step.
Fig. 1 is the schematic flow chart of active denoising method according to an embodiment of the present invention；
Fig. 2 is another schematic flow chart of active denoising method according to an embodiment of the present invention；
Fig. 3 is a kind of schematic block diagram of active noise reducing device of embodiment according to the present invention；
Fig. 4 is a kind of schematic block diagram of earphone of embodiment according to the present invention.
Specific embodiment
In order to enable the object, technical solutions and advantages of the present invention become apparent, root is described in detail below with reference to accompanying drawings
According to example embodiments of the present invention.Obviously, described embodiment is only a part of the embodiments of the present invention, rather than this hair
Bright whole embodiments, it should be appreciated that the present invention is not limited by example embodiment described herein.Based on described in the present invention
The embodiment of the present invention, those skilled in the art's obtained all other embodiment in the case where not making the creative labor
It should all fall under the scope of the present invention.
In the following description, a large amount of concrete details are given so as to provide a more thorough understanding of the present invention.So
And it is obvious to the skilled person that the present invention may not need one or more of these details and be able to
Implement.In other examples, in order to avoid confusion with the present invention, for some technical characteristics well known in the art not into
Row description.
It should be understood that the present invention can be implemented in different forms, and should not be construed as being limited to propose here
Embodiment.On the contrary, provide these embodiments will make it is open thoroughly and completely, and will fully convey the scope of the invention to
Those skilled in the art.
The purpose of term as used herein is only that description specific embodiment and not as limitation of the invention.Make herein
Used time, " one " of singular, "one" and " described/should " be also intended to include plural form, unless the context clearly indicates separately
Outer mode.It is also to be understood that term " composition " and/or " comprising ", when being used in this specification, determines the feature, whole
The presence of number, step, operations, elements, and/or components, but be not excluded for one or more other features, integer, step, operation,
The presence or addition of component, assembly unit and/or group.Herein in use, term "and/or" includes any of related listed item and institute
There is combination.
In order to thoroughly understand the present invention, detailed structure will be proposed in following description, to illustrate proposition of the present invention
Technical solution.Alternative embodiment of the invention is described in detail as follows, however other than these detailed descriptions, the present invention can be with
With other embodiments.
In order to solve aforementioned technical problem, the present invention provides a kind of active denoising method, device, earphone and computer and deposits
Storage media, wherein the described method includes: obtaining audio data；The audio number is judged using trained scene classification model
According to corresponding scene type, and export the classification results of scene type；And active noise reduction is selected according to the classification results
Mode.
Active denoising method, device, earphone and computer storage medium of the invention selects noise reduction mould according to scene type
Formula so as to avoid noise reduction bring insecurity, and avoids the triviality of manual switching noise reduction mode.
Active denoising method 100 of the invention is explained in detail and is illustrated below with reference to Fig. 1.In the premise not conflicted
Under, the feature of each embodiment of the application can be combined with each other.As shown in Figure 1, active denoising method 100 may include as follows
Step:
In step S110, audio data is obtained；
In step S120, scene type corresponding to the audio data is judged using trained scene classification model,
And export the classification results of scene type；And
In step S130, noise reduction mode is selected according to the classification results.
Illustratively, active denoising method according to an embodiment of the present invention can be in the noise reduction with memory and processor
It is realized in earphone.
Active denoising method according to an embodiment of the present invention can be deployed in earphone end, can also dispersedly be deployed in service
Device end (or cloud) and earphone end.For example, earphone end is by reception/acquisition delivery of audio data to server end (or cloud),
Active noise reduction is carried out in server end (or cloud), server end (or cloud) will be handled by active denoising method of the invention
The delivery of audio data crossed gives earphone end, and earphone end is played out according to the received processed audio data of institute.
Active denoising method according to an embodiment of the present invention can select noise reduction mode according to scene type, to avoid
Noise reduction bring insecurity, and avoid the triviality of manual switching noise reduction mode.
In one embodiment, described multiple as shown in Fig. 2, the scene type includes complex scene and non-complex scene
It include useful audio and useless audio in the audio data of miscellaneous scene.
Specifically, as shown in Fig. 2, active denoising method 200 can specifically include: firstly, collecting audio in step S210
Data；In step S220, audio frequency characteristics are extracted for the audio data；In step S230, trained scene classification is utilized
Model is classified；And in step S240, judge whether the scene type is complex scene.Wherein, when the scene class
Not Wei non-complex scene when, execute step S250, select common noise reduction mode, when the scene type be complex scene when, choosing
Select the noise reduction mode under complex scene.Further, when the scene type is complex scene, step S260 is first carried out,
Judge whether the complex scene is new complex scene, i.e., the noise whether being stored in audio database under the scene；If
The complex scene is new complex scene, thens follow the steps S270, while the noise reduction mode being switched under complex scene,
The noise in the new complex scene is acquired as audio data is expanded, is input in neural network and is trained, to scene
Disaggregated model carries out study expansion；If not new complex scene, S280 is thened follow the steps, the drop being switched under complex scene
It makes an uproar mode, but does not acquire expansion audio data.
According to embodiments of the present invention, in step S210, can be arranged at the neck ring of neck ring formula earphone and earphone several
A sound pick-up collects audio data by the sound pick-up, and extracts the noise in the audio data being collected into.
According to embodiments of the present invention, the classification results packet using trained scene classification model output scene type
It includes: feature extraction being carried out to the audio data, to obtain audio frequency characteristics；The audio frequency characteristics are inputted into the trained field
Scape disaggregated model, to obtain scene type locating for the audio data.
In one embodiment, the mode for carrying out feature extraction to the audio data can include but is not limited to Fu in short-term
In leaf transformation (STFT).Illustratively, carrying out the obtained audio frequency characteristics of feature extraction to the audio data may include frequency
Domain amplitude and/or energy information.Illustratively, carrying out the obtained audio frequency characteristics of feature extraction to the audio data can be with
Including spectral phase information.Illustratively, the obtained audio frequency characteristics of feature extraction are carried out to the audio data to be also possible to
Temporal signatures.In other examples, carrying out the obtained data characteristics of feature extraction to the audio data can also include appointing
What he can characterize the feature of the audio data.
In one embodiment, before carrying out feature extraction to the audio data, first it can be carried out at framing
Reason, and feature extraction above-mentioned is carried out frame by frame for obtaining the audio data after framing, to effectively reduce data volume, is mentioned
High treatment efficiency.
In one embodiment, in step S230, the training method of the scene classification model includes: to audio data
The original audio data that scene type has been marked in library carries out feature extraction, obtains original audio data feature and its corresponding
Type feature；And the original audio data feature and its corresponding type feature are input in deep neural network and carried out
Training, to obtain initial scene classification model.
Illustratively, the original audio data includes the sound that road surface sound and/or high frequency time repeat.Specifically
Ground, road surface sound include the audio of the reaction traffic information such as the horn for vehicle sound on different road surfaces, motor sound, brake sound, tire.It is high
The sound that the frequency repeats includes calling word, common name, the talk environment etc. that high frequency time repeats.To above-mentioned audio into
It is entered into audio database after row data collection, and marks its audio sample data.
Illustratively, the audio data characteristics of the original audio data in audio database are by manual or automatic mark
The data characteristics for the audio data that mode marks, including and be not limited in audio volume control with obvious classification characteristic part spy
Sign；The extracting method of data characteristics includes and is not limited to FFT (Fast Fourier Transformation, fast Fourier change
Change), MFCC (Mel-Frequency Cepstral Coefficient, Mel frequency cepstral coefficient) etc..
Illustratively, the feature extraction of the audio data includes carrying out feature after carrying out the audio data framing to mention
It takes.Illustratively, the method that the audio data characteristics extract include FFT, STFT, MFCC, one frame or multiframe time domain waveform or
At least one of other features of engineer.Illustratively, it includes that time domain or frequency domain are special that the audio data characteristics, which extract,
Sign.
Illustratively, the training method of the scene classification model further includes based on deep neural network, by the audio
Sample data feature obtains initial scene classification model as output training as input, corresponding type feature.Initial
Scene classification model is capable of the scene type of preliminary resolution audio frame.Since initial scene classification model is based on depth nerve
Network training, deep neural network has Episodic Memory and learning ability in initial data data basis, can pass through
It obtains expansion data and constantly carries out study expansion.
In one embodiment, the study expansion includes: the acquisition after obtaining the initial scene classification model
Audio data is expanded, and collects the associated audio data for expanding audio data；By the expansion audio data and described open up
The associated audio data of exhibition audio data, which is input in the deep neural network, to be trained, thus to the initial scene
Disaggregated model carries out study expansion, to obtain trained scene classification model.Illustratively, when recognizing new complex scene
When, acquire the related sound that the audio data in the complex scene expands audio data as expansion audio data, and described in typing
Frequency evidence, to expand out a variety of possibilities for the audio being likely to occur in the complex scene, thus in the base of original audio data
Study expansion is carried out on plinth.
Illustratively, the expansion audio data includes: the sound that road surface sound and/or high frequency time repeat.Specifically
Ground, the road surface sound that typing first appears, and learn to expand the correlation road surface sound being likely to occur；The surname that typing frequently occurs
Name, and learn to expand out corresponding pet name that may be present, English name, pseudonym etc.；High frequency in typing varying environment calls word,
And learn to expand out corresponding tone color and frequency that may be present；The different exchange scene of typing, and learn to expand out corresponding
Exchange scene that may be present etc..To sum up, the data under several scenes are combined together and expansion is carried out to neural network
It practises, subsequent ever-increasing expansion data can constantly expand scene classification model, to promote the accurate of scene Recognition
Property.
In one embodiment, the noise reduction mode under the complex scene further comprises: distinguishing having in audio data
With audio and useless audio；Active noise reduction processing is carried out to useless audio, to useful audio without active noise reduction processing.It can be with
Classified by trained disaggregated model to audio data, to distinguish useful audio therein and useless audio, such as road
The sound that face sound and/or high frequency time repeat is useful audio.The common noise reduction mode includes: to whole noises
Carry out noise reduction.
Illustratively, active noise reduction processing includes carrying out active noise reduction processing based on trained filter.Wherein, described
The training method of active noise reduction filter includes: outside measuring device to need to carry out active noise reduction region, at human ear wearing,
Filter parameter is arranged according to vocal tract transfer function in the transmission function of sound channel.Equipment acquires the outside environmental sounds of device external,
Sound can be obtained carrying out the inversion signal of the ambient sound in active noise reduction region by filter process, and reverse phase is sent out with loudspeaker
It is shot out.The arrival of device external ambient sound need to carry out active noise reduction region, at human ear wearing, with the reverse phase sound being launched
Superposition makes ambient noise remove or decay.
Active denoising method according to an embodiment of the present invention is described above exemplarily.Illustratively, according to the present invention
The active denoising method of embodiment can with memory and processor unit or system in realize.
Active denoising method according to an embodiment of the present invention can be kept away by automatically switching noise reduction mode according to scene type
Exempt from active noise reduction processing and offset insecurity brought by the sound of road surface, and avoids the cumbersome of manual switching noise reduction mode
Property.
According to another aspect of the present invention, a kind of active noise reducing device is provided.It shows with reference to Fig. 3, Fig. 3 according to this hair
A kind of schematic block diagram of active noise reducing device 300 of bright embodiment.
Active noise reducing device 300 includes that audio obtains module 310, scene classification module 320 and mode selection module 330.
The modules can execute each step/function of active denoising method 100 described above respectively.Below only to master
The major function of each module of dynamic denoising device 300 is described, and omits the detail content having been described above.
Audio obtains module 310, for obtaining audio data；
Scene classification module 320, for being judged corresponding to the audio data using trained scene classification model
Scene type, and export the classification results of scene type；And
Mode selection module 330, for selecting active noise reduction mode according to the classification results.
It may include the neck ring that neck ring formula earphone is arranged in and several sound pick-ups at earphone that audio, which obtains module 310,
Audio data is collected by the sound pick-up, and extracts the noise in the audio data being collected into.
According to embodiments of the present invention, scene classification module 320, which is configured that, carries out feature extraction to the audio data, with
Obtain audio frequency characteristics；The audio frequency characteristics are inputted into the trained scene classification model, to obtain the audio data institute
The scene type at place.
In one embodiment, the scene type includes complex scene and non-complex scene, the sound of the complex scene
Frequency includes useful audio and useless audio in.
In one embodiment, the mode for carrying out feature extraction to the audio data can include but is not limited to Fu in short-term
In leaf transformation (STFT).Illustratively, carrying out the obtained audio frequency characteristics of feature extraction to the audio data may include frequency
Domain amplitude and/or energy information.Illustratively, carrying out the obtained audio frequency characteristics of feature extraction to the audio data can be with
Including spectral phase information.Illustratively, the obtained audio frequency characteristics of feature extraction are carried out to the audio data to be also possible to
Temporal signatures.In other examples, carrying out the obtained data characteristics of feature extraction to the audio data can also include appointing
What he can characterize the feature of the audio data.
In one embodiment, before carrying out feature extraction to the audio data, first it can be carried out at framing
Reason, and feature extraction above-mentioned is carried out frame by frame for obtaining the audio data after framing, to effectively reduce data volume, is mentioned
High treatment efficiency.
In one embodiment, the training method of scene classification model used in scene classification module 320 includes: to sound
Marked in frequency database scene type original audio data carry out feature extraction, obtain original audio data feature and its
Corresponding type feature；And the original audio data feature and its corresponding type feature are input to deep neural network
In be trained, to obtain initial scene classification model.
Illustratively, the original audio data includes the sound that road surface sound and/or high frequency time repeat.Specifically
Ground, road surface sound include the audio of the reaction traffic information such as the horn for vehicle sound on different road surfaces, motor sound, brake sound, tire.It is high
The sound that the frequency repeats includes calling word, common name, the talk environment etc. that high frequency time repeats.To above-mentioned audio into
It is entered into audio database after row data collection, and marks its audio sample data.
Illustratively, the audio data characteristics of the original audio data in audio database are by manual or automatic mark
The data characteristics for the audio data that mode marks, including and be not limited in audio volume control with obvious classification characteristic part spy
Sign；The extracting method of data characteristics includes and is not limited to FFT (Fast Fourier Transformation, fast Fourier change
Change), MFCC (Mel-Frequency Cepstral Coefficient, Mel frequency cepstral coefficient) etc..
Illustratively, the feature extraction of the audio data includes carrying out feature after carrying out the audio data framing to mention
It takes.Illustratively, the method that the audio data characteristics extract include FFT, STFT, MFCC, one frame or multiframe time domain waveform or
At least one of other features of engineer.Illustratively, it includes that time domain or frequency domain are special that the audio data characteristics, which extract,
Sign.
Illustratively, the training method of the scene classification model further includes based on deep neural network, by the audio
Sample data feature obtains initial scene classification model as output training as input, corresponding type feature.Initial
Scene classification model be capable of the scene type of preliminary resolution audio frame.Since initial scene classification model is based on depth mind
Through network training, deep neural network has Episodic Memory and learning ability in initial data data basis, Ke Yitong
It crosses acquisition expansion data and constantly carries out study expansion.
In one embodiment, device 300 further includes Model Extension module, is configured to the initial scene point
Class model carries out study expansion.Specifically, after obtaining the initial scene classification model, audio data is expanded in acquisition,
And collect the associated audio data for expanding audio data；By the expansion audio data and the phase for expanding audio data
Pass audio data, which is input in the deep neural network, to be trained, thus to the initial scene classification model
It practises and expanding, to obtain trained scene classification model.Illustratively, when recognizing new complex scene, the complexity is acquired
Audio data in scene expands the associated audio data of audio data as expansion audio data, and described in typing, to expand
The a variety of possibilities for the audio being likely to occur in the complex scene out are opened up to carry out study on the basis of original audio data
Exhibition.
Illustratively, the expansion audio data includes: the sound that road surface sound and/or high frequency time repeat.Specifically
Ground, the road surface sound that typing first appears, and learn to expand the correlation road surface sound being likely to occur；The surname that typing frequently occurs
Name, and learn to expand out corresponding pet name that may be present, English name, pseudonym etc.；High frequency in typing varying environment calls word,
And learn to expand out corresponding tone color and frequency that may be present；The different exchange scene of typing, and learn to expand out corresponding
Exchange scene that may be present etc..To sum up, the data under several scenes are combined together and expansion is carried out to neural network
It practises, subsequent ever-increasing expansion data can constantly expand scene classification model, to promote the accurate of scene Recognition
Property.
In one embodiment, when the scene type is non-complex scene, mode selection module 330 is using common drop
It makes an uproar mode, when the scene type is complex scene, mode selection module 330 is using the noise reduction mode under complex scene.Tool
Body, the noise reduction mode under the complex scene further comprises: distinguishing the useful audio and useless audio in audio data；It is right
Useless audio carries out active noise reduction processing, to useful audio without active noise reduction processing.Trained classification mould can be passed through
Type classifies to audio data, to distinguish useful audio therein and useless audio, such as road surface sound and/or high frequency time weight
Existing sound of appearing again is useful audio.The common noise reduction mode includes: to carry out noise reduction to whole noises.
Illustratively, active noise reduction processing includes carrying out active noise reduction processing based on trained filter.Wherein, described
The training method of active noise reduction filter includes: outside measuring device to need to carry out active noise reduction region, at human ear wearing,
Filter parameter is arranged according to vocal tract transfer function in the transmission function of sound channel.Equipment acquires the outside environmental sounds of device external,
Sound can be obtained carrying out the inversion signal of the ambient sound in active noise reduction region by filter process, and reverse phase is sent out with loudspeaker
It is shot out.The arrival of device external ambient sound need to carry out active noise reduction region, at human ear wearing, with the reverse phase sound being launched
Superposition makes ambient noise remove or decay.
Active noise reducing device according to an embodiment of the present invention can be kept away by automatically switching noise reduction mode according to scene type
Exempt from active noise reduction processing and offset insecurity brought by the sound of road surface, and avoids the cumbersome of manual switching noise reduction mode
Property.
According to another aspect of the present invention, a kind of earphone 400 is provided.Show according to the present invention with reference to Fig. 4, Fig. 4
The schematic block diagram of the earphone 400 of embodiment.Earphone 400 can be the wireless headset of such as bluetooth, WIFI earphone, or
Wired earphone.
Earphone 400 includes storage device 410 and processor 420.Wherein, the storage of storage device 410 is for realizing basis
The program of corresponding steps in the active denoising method of the embodiment of the present invention；Processor 420 in Running storage device 410 for depositing
The program of storage, to execute the corresponding steps of active denoising method according to an embodiment of the present invention, and for realizing according to this hair
Corresponding module in the active noise reducing device of bright embodiment.
According to another aspect of the present invention, a kind of storage medium is additionally provided, stores program on said storage
Instruction, when described program instruction is run by computer or processor for executing the active denoising method of the embodiment of the present invention
Corresponding steps, and for realizing the corresponding module in active noise reducing device according to an embodiment of the present invention.The storage medium
It such as may include the storage card of smart phone, the storage unit of tablet computer, the hard disk of personal computer, read-only memory
(ROM), Erasable Programmable Read Only Memory EPROM (EPROM), portable compact disc read-only memory (CD-ROM), USB storage,
Or any combination of above-mentioned storage medium.The computer readable storage medium can be one or more computer-readable deposit
Any combination of storage media.
In one embodiment, the computer program instructions may be implemented real according to the present invention when being run by computer
Each functional module of the active noise reducing device of example is applied, and/or active noise reduction according to an embodiment of the present invention can be executed
Method.
Each module in active noise reducing device according to an embodiment of the present invention can pass through master according to an embodiment of the present invention
Computer program instructions that the processor operation of the electronic equipment of dynamic noise reduction stores in memory realize, or can be in root
The computer instruction stored in computer readable storage medium according to the computer program product of the embodiment of the present invention is by computer
It is realized when operation.
To sum up, active denoising method provided by the invention, device, earphone and computer storage medium are selected according to scene type
Noise reduction mode is selected, so as to avoid noise reduction bring insecurity, and avoids the triviality of manual switching noise reduction mode.
Although describing example embodiment by reference to attached drawing here, it should be understood that above example embodiment are only exemplary
, and be not intended to limit the scope of the invention to this.Those of ordinary skill in the art can carry out various changes wherein
And modification, it is made without departing from the scope of the present invention and spiritual.All such changes and modifications are intended to be included in appended claims
Within required the scope of the present invention.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, apparatus embodiments described above are merely indicative, for example, the division of the unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components can be tied
Another equipment is closed or is desirably integrated into, or some features can be ignored or not executed.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this specification.
Similarly, it should be understood that in order to simplify the present invention and help to understand one or more of the various inventive aspects,
To in the description of exemplary embodiment of the present invention, each feature of the invention be grouped together into sometimes single embodiment, figure,
Or in descriptions thereof.However, the method for the invention should not be construed to reflect an intention that i.e. claimed
The present invention claims features more more than feature expressly recited in each claim.More precisely, as corresponding
As claims reflect, inventive point is that all features less than some disclosed single embodiment can be used
Feature solves corresponding technical problem.Therefore, it then follows thus claims of specific embodiment are expressly incorporated in the tool
Body embodiment, wherein each, the claims themselves are regarded as separate embodiments of the invention.
It will be understood to those skilled in the art that any combination pair can be used other than mutually exclusive between feature
All features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed any method
Or all process or units of equipment are combined.Unless expressly stated otherwise, this specification (is wanted including adjoint right
Ask, make a summary and attached drawing) disclosed in each feature can be replaced with an alternative feature that provides the same, equivalent, or similar purpose.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention
Within the scope of and form different embodiments.For example, in detail in the claims, embodiment claimed it is one of any
Can in any combination mode come using.
Various component embodiments of the invention can be implemented in hardware, or to run on one or more processors
Software module realize, or be implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice
Microprocessor or digital signal processor (DSP) realize some moulds in article analytical equipment according to an embodiment of the present invention
The some or all functions of block.The present invention is also implemented as a part or complete for executing method as described herein
The program of device (for example, computer program and computer program product) in portion.It is such to realize that program of the invention can store
On a computer-readable medium, it or may be in the form of one or more signals.Such signal can be from internet
Downloading obtains on website, is perhaps provided on the carrier signal or is provided in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability
Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch
To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame
Claim.
The above description is merely a specific embodiment or to the explanation of specific embodiment, protection of the invention
Range is not limited thereto, and anyone skilled in the art in the technical scope disclosed by the present invention, can be easily
Expect change or replacement, should be covered by the protection scope of the present invention.Protection scope of the present invention should be with claim
Subject to protection scope.
Claims (10)
1. a kind of active denoising method, which is characterized in that the described method includes:
Obtain audio data；
Scene type corresponding to the audio data is judged using trained scene classification model, and exports scene type
Classification results；And
Active noise reduction mode is selected according to the classification results.
2. active denoising method as described in claim 1, which is characterized in that the training method packet of the scene classification model
It includes:
Feature extraction is carried out to the original audio data for having marked scene type in audio database, obtains original audio data
Feature and its corresponding type feature；
The original audio data feature and its corresponding type feature are input in deep neural network and are trained to obtain
To initial scene classification model.
3. active denoising method as claimed in claim 2, which is characterized in that the training method of the scene classification model is also wrapped
It includes:
After obtaining the initial scene classification model, audio data is expanded in acquisition, and collects the expansion audio data
Associated audio data；
The expansion audio data and the associated audio data for expanding audio data are input to the deep neural network
In be trained, to carry out study expansion to the initial scene classification model, to obtain the trained scene point
Class model.
4. active denoising method as claimed in claim 3, which is characterized in that the original audio data and/or the expansion
Audio data includes: the sound that road surface sound and/or high frequency time repeat.
5. active denoising method as described in claim 1, which is characterized in that the scene type includes complex scene and non-multiple
Miscellaneous scene, it is described to include: according to classification results selection noise reduction mode
When the scene type is non-complex scene, common noise reduction mode is selected, to carry out noise reduction to whole noises；
When the scene type is complex scene, the noise reduction mode under complex scene is selected.
6. active denoising method as claimed in claim 5, which is characterized in that the noise reduction mode under the complex scene includes:
Distinguish the useful audio and useless audio in the audio data；
Active noise reduction processing is carried out to the useless audio, to the useful audio without active noise reduction processing.
7. active denoising method as described in claim 1, which is characterized in that described to be sentenced using trained scene classification model
Break scene type corresponding to the audio data, and the classification results for exporting scene type include:
Feature extraction is carried out to the audio data, to obtain audio frequency characteristics；
The audio frequency characteristics are inputted into the trained scene classification model, to obtain the classification results of the scene type.
8. a kind of active noise reducing device characterized by comprising
Audio obtains module, for obtaining audio data；
Scene classification module, for judging scene class corresponding to the audio data using trained scene classification model
Not, and the classification results of scene type are exported；And
Mode selection module, for selecting active noise reduction mode according to the classification results.
9. a kind of earphone, including memory, processor and the meter for being stored on the memory and running on the processor
Calculation machine program, which is characterized in that the processor realizes any one of claims 1 to 7 the method when executing described program
The step of.
10. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that described program is held by processor
The step of any one of claims 1 to 7 the method is realized when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811407883.2A CN109346055A (en) | 2018-11-23 | 2018-11-23 | Active denoising method, device, earphone and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811407883.2A CN109346055A (en) | 2018-11-23 | 2018-11-23 | Active denoising method, device, earphone and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109346055A true CN109346055A (en) | 2019-02-15 |
Family
ID=65317350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811407883.2A Pending CN109346055A (en) | 2018-11-23 | 2018-11-23 | Active denoising method, device, earphone and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109346055A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110191388A (en) * | 2019-05-31 | 2019-08-30 | 深圳市荣盛智能装备有限公司 | Bone conduction earphone noise-reduction method, device, electronic equipment and storage medium |
| CN110809211A (en) * | 2020-01-08 | 2020-02-18 | 恒玄科技(北京)有限公司 | Method for actively reducing noise of earphone, active noise reduction system and earphone |
| CN110972014A (en) * | 2019-12-11 | 2020-04-07 | 歌尔智能科技有限公司 | Parameter adjustment method and device for active noise reduction earphone and wireless earphone |
| CN111833895A (en) * | 2019-04-23 | 2020-10-27 | 北京京东尚科信息技术有限公司 | Audio signal processing method, apparatus, computer device and medium |
| US10834494B1 (en) | 2019-12-13 | 2020-11-10 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| CN112767908A (en) * | 2020-12-29 | 2021-05-07 | 安克创新科技股份有限公司 | Active noise reduction method based on key sound recognition, electronic equipment and storage medium |
| CN112767965A (en) * | 2019-11-01 | 2021-05-07 | 上海博泰悦臻电子设备制造有限公司 | Method, system, medium, and service/terminal for generating/applying noise recognition model |
| WO2021114514A1 (en) * | 2019-12-13 | 2021-06-17 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| CN113160789A (en) * | 2021-03-05 | 2021-07-23 | 南京每深智能科技有限责任公司 | Active noise reduction device and method |
| CN113259824A (en) * | 2021-05-14 | 2021-08-13 | 谷芯(广州)技术有限公司 | Real-time multi-channel digital hearing aid noise reduction method and system |
| CN113810814A (en) * | 2021-08-17 | 2021-12-17 | 百度在线网络技术（北京）有限公司 | Earphone mode switching control method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104616662A (en) * | 2015-01-27 | 2015-05-13 | 中国科学院理化技术研究所 | Active noise reduction method and device |
| CN106205609A (en) * | 2016-07-05 | 2016-12-07 | 山东师范大学 | A kind of based on audio event and the audio scene recognition method of topic model and device |
| CN106572411A (en) * | 2016-09-29 | 2017-04-19 | 乐视控股（北京）有限公司 | Noise cancelling control method and relevant device |
| US20180254033A1 (en) * | 2016-11-01 | 2018-09-06 | Davi Audio | Smart Noise Reduction System and Method for Reducing Noise |
| CN108764304A (en) * | 2018-05-11 | 2018-11-06 | Oppo广东移动通信有限公司 | scene recognition method, device, storage medium and electronic equipment |
-
2018
- 2018-11-23 CN CN201811407883.2A patent/CN109346055A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104616662A (en) * | 2015-01-27 | 2015-05-13 | 中国科学院理化技术研究所 | Active noise reduction method and device |
| CN106205609A (en) * | 2016-07-05 | 2016-12-07 | 山东师范大学 | A kind of based on audio event and the audio scene recognition method of topic model and device |
| CN106572411A (en) * | 2016-09-29 | 2017-04-19 | 乐视控股（北京）有限公司 | Noise cancelling control method and relevant device |
| US20180254033A1 (en) * | 2016-11-01 | 2018-09-06 | Davi Audio | Smart Noise Reduction System and Method for Reducing Noise |
| CN108764304A (en) * | 2018-05-11 | 2018-11-06 | Oppo广东移动通信有限公司 | scene recognition method, device, storage medium and electronic equipment |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111833895B (en) * | 2019-04-23 | 2023-12-05 | 北京京东尚科信息技术有限公司 | Audio signal processing method, device, computer equipment and medium |
| CN111833895A (en) * | 2019-04-23 | 2020-10-27 | 北京京东尚科信息技术有限公司 | Audio signal processing method, apparatus, computer device and medium |
| CN110191388A (en) * | 2019-05-31 | 2019-08-30 | 深圳市荣盛智能装备有限公司 | Bone conduction earphone noise-reduction method, device, electronic equipment and storage medium |
| CN112767965A (en) * | 2019-11-01 | 2021-05-07 | 上海博泰悦臻电子设备制造有限公司 | Method, system, medium, and service/terminal for generating/applying noise recognition model |
| CN110972014A (en) * | 2019-12-11 | 2020-04-07 | 歌尔智能科技有限公司 | Parameter adjustment method and device for active noise reduction earphone and wireless earphone |
| CN110972014B (en) * | 2019-12-11 | 2022-03-01 | 歌尔智能科技有限公司 | Parameter adjustment method and device for active noise reduction earphone and wireless earphone |
| US11595748B2 (en) | 2019-12-13 | 2023-02-28 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| US11653141B2 (en) | 2019-12-13 | 2023-05-16 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| WO2021114514A1 (en) * | 2019-12-13 | 2021-06-17 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| US11330359B2 (en) | 2019-12-13 | 2022-05-10 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| US11317192B2 (en) | 2019-12-13 | 2022-04-26 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| US10834494B1 (en) | 2019-12-13 | 2020-11-10 | Bestechnic (Shanghai) Co., Ltd. | Active noise control headphones |
| CN111541971A (en) * | 2020-01-08 | 2020-08-14 | 恒玄科技(北京)有限公司 | Method for actively reducing noise of earphone, active noise reduction system and earphone |
| CN110809211A (en) * | 2020-01-08 | 2020-02-18 | 恒玄科技(北京)有限公司 | Method for actively reducing noise of earphone, active noise reduction system and earphone |
| CN112767908A (en) * | 2020-12-29 | 2021-05-07 | 安克创新科技股份有限公司 | Active noise reduction method based on key sound recognition, electronic equipment and storage medium |
| CN112767908B (en) * | 2020-12-29 | 2024-05-21 | 安克创新科技股份有限公司 | Active noise reduction method based on key voice recognition, electronic equipment and storage medium |
| CN113160789A (en) * | 2021-03-05 | 2021-07-23 | 南京每深智能科技有限责任公司 | Active noise reduction device and method |
| CN113259824A (en) * | 2021-05-14 | 2021-08-13 | 谷芯(广州)技术有限公司 | Real-time multi-channel digital hearing aid noise reduction method and system |
| CN113810814A (en) * | 2021-08-17 | 2021-12-17 | 百度在线网络技术（北京）有限公司 | Earphone mode switching control method and device, electronic equipment and storage medium |
| CN113810814B (en) * | 2021-08-17 | 2023-12-01 | 百度在线网络技术（北京）有限公司 | Earphone mode switching control method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109346055A (en) | Active denoising method, device, earphone and computer storage medium | |
| Mesaros et al. | A multi-device dataset for urban acoustic scene classification | |
| CN106504768B (en) | Phone testing audio frequency classification method and device based on artificial intelligence | |
| CN110246490A (en) | Voice keyword detection method and relevant apparatus | |
| CN108597498A (en) | Multi-microphone voice acquisition method and device | |
| CN107360387A (en) | Video recording method and device and terminal equipment | |
| CN111640411B (en) | Audio synthesis method, device and computer readable storage medium | |
| CN106486130A (en) | Noise elimination, audio recognition method and device | |
| CN109935226A (en) | A kind of far field speech recognition enhancing system and method based on deep neural network | |
| CN111868823B (en) | Sound source separation method, device and equipment | |
| CN111385688A (en) | Active noise reduction method, device and system based on deep learning | |
| CN112382300A (en) | Voiceprint identification method, model training method, device, equipment and storage medium | |
| CN107293305A (en) | It is a kind of to improve the method and its device of recording quality based on blind source separation algorithm | |
| CN108847221A (en) | Audio recognition method, device, storage medium and electronic equipment | |
| CN110910876A (en) | Article sound searching device and control method, and voice control setting method and system | |
| CN110232909A (en) | A kind of audio-frequency processing method, device, equipment and readable storage medium storing program for executing | |
| CN109308900A (en) | Headphone device, speech processing system and method for speech processing | |
| CN111142066A (en) | Direction-of-arrival estimation method, server, and computer-readable storage medium | |
| CN112382302A (en) | Baby cry identification method and terminal equipment | |
| CN110718229A (en) | Detection method for record playback attack and training method corresponding to detection model | |
| Diaconita et al. | Do you hear what i hear? using acoustic probing to detect smartphone locations | |
| US20230276165A1 (en) | Audio signal processing method, terminal device and storage medium | |
| CN113327631A (en) | Emotion recognition model training method, emotion recognition method and emotion recognition device | |
| CN109686359A (en) | Speech output method, terminal and computer readable storage medium | |
| CN110556114B (en) | Speaker identification method and device based on attention mechanism |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190215 |