CN106486130A - Noise elimination, audio recognition method and device - Google Patents
Noise elimination, audio recognition method and device Download PDFInfo
- Publication number
- CN106486130A CN106486130A CN201510524909.1A CN201510524909A CN106486130A CN 106486130 A CN106486130 A CN 106486130A CN 201510524909 A CN201510524909 A CN 201510524909A CN 106486130 A CN106486130 A CN 106486130A
- Authority
- CN
- China
- Prior art keywords
- voice
- vocal print
- data
- original audio
- audio data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000008030 elimination Effects 0.000 title claims abstract description 26
- 238000003379 elimination reaction Methods 0.000 title claims abstract description 26
- 230000001755 vocal effect Effects 0.000 claims abstract description 128
- 238000012545 processing Methods 0.000 claims description 21
- 241000208340 Araliaceae Species 0.000 claims 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 4
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 4
- 235000008434 ginseng Nutrition 0.000 claims 4
- 230000009467 reduction Effects 0.000 abstract description 16
- 230000001629 suppression Effects 0.000 abstract description 6
- 230000008859 change Effects 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 9
- 238000010168 coupling process Methods 0.000 description 7
- 238000005859 coupling reaction Methods 0.000 description 7
- 230000008878 coupling Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/20—Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
Landscapes
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
- Telephone Function (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The embodiment of the present invention provides a kind of noise elimination, audio recognition method and device.Noise cancellation method is by being based on specific vocal print parameter, voice print matching is carried out to acquired pending original audio data, make it possible to the voice print matching result according to the voice print matching, effective voice data is obtained from the pending original audio data, other voice signals such as noise signal etc. is gathered without the need for extra voice collection device, can avoid as the distance between the corresponding signal source of voice signal and two microphones change problem voice signal carried out with the suppression of noise signal equal extent for causing in prior art, so as to improve the reliability of noise reduction, while the sound quality after noise reduction can be effectively improved.
Description
【Technical field】
The present invention relates to noise management technique, more particularly to a kind of noise elimination, audio recognition method and dress
Put.
【Background technology】
As sound processing techniques development is increasingly faster, terminal-pair needs the requirement of sound quality to be processed
More and more higher, noise reduction technology are arisen at the historic moment.Current noise reduction technology is mainly actively dropped using dual microphone
Make an uproar, by certain algorithm, the voice data collected using a microphone (corresponds to noise signal
The voice signal weaker with signal strength signal intensity) voice data that another microphone is collected (is corresponded to
The stronger voice signal of noise signal and signal strength signal intensity), carry out noise suppressed process.
If however, the corresponding signal source of voice signal is (for example：The face of people) with two microphones it
Between distance change, voice signal may be caused to be confirmed as noise so that also enter voice signal
The suppression that has gone with noise signal equal extent, the sound quality having had a strong impact on after noise reduction, result in
The reduction of the reliability of noise reduction.
【Content of the invention】
The many aspects of the present invention provide a kind of noise elimination, audio recognition method and device, in order to improve
The reliability of noise reduction.
An aspect of of the present present invention, provides a kind of noise cancellation method, including：
Based on specific vocal print parameter, voice print matching is carried out to acquired pending original audio data；
According to the voice print matching result of the voice print matching, obtain from the pending original audio data
Effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
The vocal print parameter that specific vocal print parameter is targeted customer is stated, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is obtained from the pending original audio data, as described
Effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation,
Described based on specific vocal print parameter, before carrying out voice print matching to acquired pending original audio data,
Methods described also includes：
Obtain the voice signal of the targeted customer；
Based on the voice signal of the acquired targeted customer, the vocal print parameter of the targeted customer is obtained.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
The vocal print parameter that specific vocal print parameter is the noise signal of target environment is stated, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is removed from the pending original audio data, as described
Effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation,
Described based on specific vocal print parameter, before carrying out voice print matching to acquired pending original audio data,
Methods described also includes：
Obtain the noise signal of the target environment；
Based on the noise signal of the acquired target environment, the vocal print parameter of the noise signal is obtained.
Another aspect of the present invention, provides a kind of noise elimination apparatus, including：
Voice print matching unit, for being based on specific vocal print parameter, to acquired pending original audio number
According to carrying out voice print matching；
Effective voice data acquiring unit, for the voice print matching result according to the voice print matching, from institute
State and in pending original audio data, obtain effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
The vocal print parameter that specific vocal print parameter is targeted customer is stated, and
Effective voice data acquiring unit is used for obtaining vocal print from the pending original audio data
The voice data that the match is successful, used as effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
Stating noise elimination apparatus also includes：
Voice signal acquiring unit, for obtaining the voice signal of the targeted customer；
First vocal print gain of parameter unit, for the voice signal based on the acquired targeted customer,
Obtain the vocal print parameter of the targeted customer.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
The vocal print parameter that specific vocal print parameter is the noise signal of target environment is stated, and
Effective voice data acquiring unit is used for removing vocal print from the pending original audio data
The voice data that the match is successful, used as effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
Stating noise elimination apparatus also includes：
Noise signal acquiring unit, for obtaining the noise signal of the target environment；
Second vocal print gain of parameter unit, for the noise signal based on the acquired target environment,
Obtain the vocal print parameter of the noise signal.
Another aspect of the present invention, provides a kind of audio recognition method, including：
Obtain pending original audio data；
Based on specific vocal print parameter, voice print matching is carried out to acquired described pending original audio data；
According to the voice print matching result of the voice print matching, obtain from the pending original audio data
Effective voice data；
Voice recognition processing is carried out to effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
The vocal print parameter that specific vocal print parameter is targeted customer is stated, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is obtained from the pending original audio data, as described
Effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation,
Described based on specific vocal print parameter, voice print matching is carried out to acquired described pending original audio data
Before, methods described also includes：
Obtain the voice signal of the targeted customer；
Based on the voice signal of the acquired targeted customer, the vocal print parameter of the targeted customer is obtained.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation, institute
The vocal print parameter that specific vocal print parameter is the noise signal of target environment is stated, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is removed from the pending original audio data, as described
Effective voice data.
Aspect as above and arbitrary possible implementation, it is further provided a kind of implementation,
Described based on specific vocal print parameter, voice print matching is carried out to acquired described pending original audio data
Before, methods described also includes：
Obtain the noise signal of the target environment；
Based on the noise signal of the acquired target environment, the vocal print parameter of the noise signal is obtained.
Another aspect of the present invention, provides a kind of speech recognition equipment, including：
Original audio data acquiring unit, for obtaining pending original audio data；
Noise elimination apparatus as above；
Voice recognition unit, for carrying out voice recognition processing to effective voice data.
As shown from the above technical solution, on the one hand, the embodiment of the present invention by be based on specific vocal print parameter,
Voice print matching is carried out to acquired pending original audio data, enabling according to the voice print matching
Voice print matching result, obtain effective voice data from the pending original audio data, without the need for volume
Outer voice collection device gathers other voice signals such as noise signal etc., can avoid in prior art
Due to the distance between the corresponding signal source of voice signal and two microphones change cause to voice
Signal carries out the problem with the suppression of noise signal equal extent, so as to improve the reliability of noise reduction, with
When can effectively improve sound quality after noise reduction.
As shown from the above technical solution, on the other hand, the embodiment of the present invention is by obtaining pending original sound
Frequency evidence, and then specific vocal print parameter is based on, acquired described pending original audio data is carried out
Voice print matching, enabling according to the voice print matching result of the voice print matching, from described pending original
Effective voice data is obtained in voice data, and voice recognition processing is carried out to effective voice data,
As effective voice data is according to specific vocal print parameter, extract in pending original audio data
Voice data, this portion of audio data may be considered the voice signal of the user of using terminal, therefore,
No longer include noise signal in effective voice data, so as to effectively increase sound quality.
In addition, the technical scheme provided using the present invention, it is only necessary to a voice collection device, Neng Gouyou
Effect reduces cost.
【Description of the drawings】
For the technical scheme being illustrated more clearly that in the embodiment of the present invention, below will be to embodiment or existing
Accompanying drawing to be used needed for technology description is briefly described, it should be apparent that, in describing below
Accompanying drawing is some embodiments of the present invention, for those of ordinary skill in the art, is not paying creation
Property laborious on the premise of, other accompanying drawings can also be obtained according to these accompanying drawings.
The schematic flow sheet of the noise cancellation method that Fig. 1 is provided for one embodiment of the invention；
Fig. 2 is the situation of the vocal print parameter of targeted customer for specific vocal print parameter in the corresponding embodiment of Fig. 1
Under noise cancellation method schematic flow sheet；
Fig. 3 is the vocal print of the noise signal of target environment for specific vocal print parameter in the corresponding embodiment of Fig. 1
The schematic flow sheet of the noise cancellation method in the case of parameter；
The schematic flow sheet of the audio recognition method that Fig. 4 is provided for another embodiment of the present invention；
The structural representation of the noise elimination apparatus that Fig. 5 is provided for another embodiment of the present invention；
Fig. 6 is the situation of the vocal print parameter of targeted customer for specific vocal print parameter in the corresponding embodiment of Fig. 5
Under noise elimination apparatus structural representation；
Fig. 7 is the vocal print of the noise signal of target environment for specific vocal print parameter in the corresponding embodiment of Fig. 5
The structural representation of the noise elimination apparatus in the case of parameter；
The structural representation of the speech recognition equipment that Fig. 8 is provided for another embodiment of the present invention.
【Specific embodiment】
Purpose, technical scheme and advantage for making the embodiment of the present invention is clearer, below in conjunction with this
Accompanying drawing in bright embodiment, is clearly and completely described to the technical scheme in the embodiment of the present invention,
Obviously, described embodiment is a part of embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained under the premise of creative work is not made
The every other embodiment for obtaining, belongs to the scope of protection of the invention.
It should be noted that involved terminal in the embodiment of the present invention can include but is not limited to mobile phone,
Personal digital assistant (Personal Digital Assistant, PDA), radio hand-held equipment, flat board electricity
Brain (Tablet Computer), PC (Personal Computer, PC), MP3 are played
Device, MP4 player, wearable device (for example, intelligent glasses, intelligent watch, Intelligent bracelet etc.)
Deng.
In addition, the terms "and/or", only a kind of incidence relation of description affiliated partner, represents
There may be three kinds of relations, for example, A and/or B, can represent：, while there is A in individualism A
And these three situations of B, individualism B.In addition, character "/" herein, typicallys represent forward-backward correlation pair
As if a kind of relation of "or".
The schematic flow sheet of the noise cancellation method that Fig. 1 is provided for one embodiment of the invention, as shown in Figure 1.
101st, specific vocal print parameter is based on, voice print matching is carried out to acquired pending original audio data.
102nd, according to the voice print matching result of the voice print matching, from the pending original audio data
Obtain effective voice data.
It should be noted that 101～102 executive agent can be partly or entirely to be located locally terminal
Application, or can also be the plug-in unit or SDK being located locally in the application of terminal
Functional units such as (Software Development Kit, SDK), or can also be positioned at network side
Server in process engine, or can also be the distributed system positioned at network side, the present embodiment
This is not particularly limited.
It is understood that the application can be mounted in the local program in terminal (nativeApp),
Or can also be a web page program (webApp) of browser in terminal, the present embodiment to this not
It is particularly limited.
So, by specific vocal print parameter is based on, sound is carried out to acquired pending original audio data
Line mates, enabling according to the voice print matching result of the voice print matching, from the pending original sound
Effective voice data is obtained in frequency evidence, other voice signal examples are gathered without the need for extra voice collection device
Such as noise signal etc., can avoid due to the corresponding signal source of voice signal and two Mikes in prior art
The distance between wind changes voice signal being carried out and the suppression of noise signal equal extent of causing
Problem, so as to improve the reliability of noise reduction, while the sound quality after noise reduction can be effectively improved.
In the present invention, it is possible to use voice collection device, the pending original audio data is obtained.Its
In, the voice collection device, can be terminal built-in or external microphone etc., and the present embodiment is to this
It is not particularly limited.
Specifically voice collection device can be utilized, gather needs the sound of voice signal to be processed comprising terminal
Signal.Generally, may adulterate in the voice signal noise signal.Then, then will can be gathered
The voice signal is converted to pending original audio data.
Specifically, so-called pending original audio data, be by the numeral converted by audio signal
Signal.For example, specifically the voice signal can be sampled, quantify and coded treatment, to obtain
Pulse code modulation (Pulse Code Modulation, PCM) data, used as pending original audio
Data.
In the present embodiment, without the need for additionally gathering the voice data of auxiliary using extra voice collection device again,
And only need to gather pending original audio data using a voice collection device, effectively can drop
Low cost.
Alternatively, in a possible implementation of the present embodiment, in 101, specifically can be right
The pending original audio data carries out sub-frame processing, to obtain at least frame data, and then at least
Audio analysis process is carried out per frame data in one frame data, to obtain the vocal print feature of every frame data.Then,
Based on specific vocal print parameter, the vocal print feature to the pending original audio data carries out matching treatment.
The two is consistent, then the match is successful for explanation, and the two is inconsistent, then it fails to match for explanation.
So-called consistent, may refer to all unanimously i.e. completely the same, or it is consistent to can also refer to part,
The present embodiment is not particularly limited to this.
Specifically, can to the pending original audio data according to prefixed time interval, for example,
20ms, carries out the data overlap for having part between sub-frame processing, and consecutive frame, such as 50% data weight
Folded, in such manner, it is possible to obtain at least frame data of the pending original audio data.
So-called vocal print feature, is the exclusive feature of a voice data, refers to represent a section audio
The digital signature based on content of the important acoustic feature of data, its main purpose are to set up a kind of effective machine
The perception acoustical quality for making to compare two voice datas.Note it not being directly relatively generally very bigger here
Voice data itself, but compare its corresponding generally less vocal print feature.
During a concrete implementation, the vocal print feature, can include but is not limited to the mankind's
The relevant acoustic feature of the anatomical structure of pronunciation mechanism, for example, frequency spectrum, cepstrum, formant, base
Sound, reflectance factor, etc..
Alternatively, in a possible implementation of the present embodiment, before 101, can also enter
One step arranges the specific vocal print parameter, in order to the basic parameter as voice print matching.Specifically, institute
State the vocal print parameter that specific vocal print parameter can be targeted customer, or the noise that can also be target environment
The vocal print parameter of signal, the present embodiment are not particularly limited to this.Respectively in detail two kinds are described below
Specific vocal print parameter is respectively the vocal print parameter of the noise signal of the vocal print parameter and target environment of targeted customer
When, the noise cancellation method provided by the present embodiment.
Fig. 2 is the situation of the vocal print parameter of targeted customer for specific vocal print parameter in the corresponding embodiment of Fig. 1
Under noise cancellation method schematic flow sheet, as shown in Figure 2.
201st, the vocal print parameter based on targeted customer, carries out sound to acquired pending original audio data
Line mates.
Alternatively, in a possible implementation of the present embodiment, before 201, can also enter
One step obtains the voice signal of the targeted customer, and then, then can be used based on the acquired target
The voice signal at family, obtains the vocal print parameter of the targeted customer.
Specifically, specifically the voice signal of targeted customer can be sampled, quantify and coded treatment,
To obtain PCM data, as audio user data.Then, then the audio user data can be entered
Row sub-frame processing, to obtain at least frame data, and then to carrying out sound per frame data at least frame data
Frequency analysis are processed, to obtain the vocal print parameter of every frame data.
For example, it is possible to the audio user data according to prefixed time interval, for example, 20ms, carry out
There is the data overlap of part between sub-frame processing, and consecutive frame, such as 50% data overlap, so,
It is obtained in that at least frame data of the audio user data.
202nd, voice print matching successful voice data is obtained from the pending original audio data, as
Effective voice data.
In this implementation, the specific vocal print parameter refers to the voice signal according to targeted customer,
The vocal print parameter of the voice signal of the targeted customer for being obtained.Therefore, the vocal print feature that the match is successful,
Vocal print feature voice signal that the targeted customer of using terminal sent corresponding to then may be considered.
Fig. 3 is the vocal print of the noise signal of target environment for specific vocal print parameter in the corresponding embodiment of Fig. 1
The schematic flow sheet of the noise cancellation method in the case of parameter, as shown in Figure 3.
301st, the vocal print parameter based on the noise signal of target environment, to acquired pending original audio
Data carry out voice print matching.
Alternatively, in a possible implementation of the present embodiment, before 301, can also enter
One step obtains the noise signal of the target environment, and then, then can be based on the acquired target ring
The noise signal in border, obtains the vocal print parameter of the noise signal.
Specifically, specifically the voice signal of target environment can be sampled, quantify and coded treatment,
To obtain PCM data, as environmental audio data.Then, then the environmental audio data can be entered
Row sub-frame processing, to obtain at least frame data, and then to carrying out sound per frame data at least frame data
Frequency analysis are processed, to obtain the vocal print parameter of every frame data.
For example, it is possible to the environmental audio data according to prefixed time interval, for example, 20ms, carry out
There is the data overlap of part between sub-frame processing, and consecutive frame, such as 50% data overlap, so,
It is obtained in that at least frame data of the environmental audio data.
302nd, voice print matching successful voice data is removed from the pending original audio data, as
Effective voice data.
In this implementation, the specific vocal print parameter refers to the noise signal according to target environment,
The vocal print parameter of the noise signal of the target environment for being obtained.Therefore, the vocal print feature that the match is successful,
Vocal print feature noise signal in the target environment at terminal place produced by corresponding to then may be considered.
It is understood that in addition to the specific vocal print parameter of above two, can also be by least one warp
Parameter is tested, as the specific vocal print parameter.
It should be noted that after the specific vocal print parameter is obtained, in addition it is also necessary to will be obtained further
Specific vocal print parameter carry out storage process.Specifically, specifically can by the specific vocal print parameter for being obtained,
Stored in the storage device of terminal.
During a concrete implementation, the storage device of the terminal can be slow storage device,
The hard disk of computer system is specifically as follows, or can also be physical memory for the inoperative internal memory of mobile phone,
For example, read-only storage (Read-Only Memory, ROM) and RAM card etc., the present embodiment is to this
It is not particularly limited.
During another concrete implementation, the storage device of the terminal can also be set for quick storage
Standby, it is specifically as follows the internal memory of computer system, or for the running memory of mobile phone can also is in system
Deposit, for example, random access memory (Random Access Memory, RAM) etc., the present embodiment pair
This is not particularly limited.
Alternatively, in a possible implementation of the present embodiment, after 102, can also enter
One step carries out voice recognition processing to effective voice data.
As effective voice data is according to specific vocal print parameter, extract in pending original audio data
Voice data out, this portion of audio data may be considered the voice signal of the user of using terminal,
Therefore, no longer include noise signal in effective voice data, so as to effectively increase sound quality.
Further, voice recognition processing is carried out to effective voice data, the recognition result for being obtained,
Accuracy rate is higher.
In the present embodiment, by being based on specific vocal print parameter, to acquired pending original audio data
Carry out voice print matching, enabling according to the voice print matching result of the voice print matching, from described pending
Effective voice data is obtained in original audio data, gathers other sound without the need for extra voice collection device
Signal such as noise signal etc., can avoid due to the corresponding signal source of voice signal and two in prior art
The distance between individual microphone changes voice signal being carried out and noise signal equal extent of causing
The problem of suppression, so as to improve the reliability of noise reduction, while the sound after noise reduction can be effectively improved
Quality.
In addition, the technical scheme provided using the present invention, it is only necessary to a voice collection device, Neng Gouyou
Effect reduces cost.
The schematic flow sheet of the audio recognition method that Fig. 4 is provided for another embodiment of the present invention, as Fig. 4 institute
Show.
401st, pending original audio data is obtained.
402nd, specific vocal print parameter is based on, vocal print is carried out to acquired described pending original audio data
Coupling.
403rd, according to the voice print matching result of the voice print matching, from the pending original audio data
Obtain effective voice data.
404th, voice recognition processing is carried out to effective voice data.
It should be noted that 401～404 executive agent can be partly or entirely to be located locally terminal
Application, or can also be the plug-in unit or SDK being located locally in the application of terminal
Functional units such as (Software Development Kit, SDK), or can also be positioned at network side
Server in process engine, or can also be the distributed system positioned at network side, the present embodiment
This is not particularly limited.
It is understood that the application can be mounted in the local program in terminal (nativeApp),
Or can also be a web page program (webApp) of browser in terminal, the present embodiment to this not
It is particularly limited.
In the present invention, 402 and 403 detailed description, may refer to the embodiment corresponding to Fig. 1～Fig. 3
In related content, here is omitted.
In the present embodiment, by pending original audio data is obtained, and then specific vocal print parameter is based on,
Voice print matching is carried out to acquired described pending original audio data, enabling according to the vocal print
The voice print matching result of coupling, obtains effective voice data from the pending original audio data, and
Voice recognition processing is carried out to effective voice data, due to effective voice data be according to specific vocal print
Parameter, the voice data extracted in pending original audio data, this portion of audio data are permissible
It is considered the voice signal of the user of using terminal, therefore, in effective voice data, no longer includes noise
Signal, so as to effectively increase sound quality.
Further, voice recognition processing is carried out to effective voice data, the recognition result for being obtained,
Accuracy rate is higher.
In addition, the technical scheme provided using the present invention, it is only necessary to a voice collection device, Neng Gouyou
Effect reduces cost.
It should be noted that for aforesaid each method embodiment, in order to be briefly described, therefore by its all table
State as a series of combination of actions, but those skilled in the art should know, the present invention is not by being retouched
The restriction of the sequence of movement that states, because according to the present invention, some steps can be using other orders or same
Shi Jinhang.Secondly, those skilled in the art should also know, embodiment described in this description all belongs to
In preferred embodiment, necessary to involved action and the module not necessarily present invention.
In the above-described embodiments, the description to each embodiment all emphasizes particularly on different fields, and does not have in certain embodiment
The part of detailed description, may refer to the associated description of other embodiment.
The structural representation of the noise elimination apparatus that Fig. 5 is provided for another embodiment of the present invention, as Fig. 5 institute
Show.The noise elimination apparatus of the present embodiment can include that voice print matching unit 51 and effective voice data are obtained
Unit 52.Wherein, voice print matching unit 51, for being based on specific vocal print parameter, waits to locate to acquired
Reason original audio data carries out voice print matching；Effective voice data acquiring unit 52, for according to the sound
The voice print matching result of line coupling, obtains effective voice data from the pending original audio data.
It should be noted that the noise elimination apparatus that provided of the present embodiment can be partly or entirely position
In the application of local terminal, or can also be the plug-in unit or software development being located locally in the application of terminal
The functional units such as kit (Software Development Kit, SDK), or can also be positioned at
Process engine in the server of network side, or can also be the distributed system positioned at network side, this
Embodiment is not particularly limited to this.
It is understood that the application can be mounted in the local program in terminal (nativeApp),
Or can also be a web page program (webApp) of browser in terminal, the present embodiment to this not
It is particularly limited.
Alternatively, in a possible implementation of the present embodiment, the specific vocal print parameter is mesh
The vocal print parameter of mark user；Correspondingly, effective voice data acquiring unit 52, specifically can be used for
The successful voice data of voice print matching is obtained from the pending original audio data, as described effective
Voice data.
Alternatively, in a possible implementation of the present embodiment, as shown in fig. 6, the present embodiment
The noise elimination apparatus for being provided can further include：
Voice signal acquiring unit 61, for obtaining the voice signal of the targeted customer；
First vocal print gain of parameter unit 62, for the voice signal based on the acquired targeted customer,
Obtain the vocal print parameter of the targeted customer.
Alternatively, in a possible implementation of the present embodiment, the specific vocal print parameter is mesh
The vocal print parameter of the noise signal of mark environment；Correspondingly, effective voice data acquiring unit 52, tool
Body can be used for removing the successful voice data of voice print matching from the pending original audio data, make
For effective voice data.
Alternatively, in a possible implementation of the present embodiment, as shown in fig. 7, the present embodiment
The noise elimination apparatus for being provided can further include：
Noise signal acquiring unit 71, for obtaining the noise signal of the target environment；
Second vocal print gain of parameter unit 72, for the noise signal based on the acquired target environment,
Obtain the vocal print parameter of the noise signal.
It should be noted that method in the corresponding embodiment of Fig. 1～Fig. 3, all can be provided by the present embodiment
Noise elimination apparatus realize.Describe the phase that may refer in the corresponding embodiment of Fig. 1～Fig. 3 in detail inside the Pass
Hold, here is omitted.
In the present embodiment, specific vocal print parameter is based on by voice print matching unit, to acquired pending
Original audio data carries out voice print matching so that effectively voice data acquiring unit can be according to the vocal print
The voice print matching result of coupling, obtains effective voice data, no from the pending original audio data
Need extra voice collection device that other voice signals such as noise signal etc. is gathered, existing skill can be avoided
In art due to the distance between the corresponding signal source of voice signal and two microphones change cause right
Voice signal carries out the problem with the suppression of noise signal equal extent, so as to improve the reliability of noise reduction,
While the sound quality after noise reduction can be effectively improved.
In addition, the technical scheme provided using the present invention, it is only necessary to a voice collection device, Neng Gouyou
Effect reduces cost.
The structural representation of the speech recognition equipment that Fig. 8 is provided for another embodiment of the present invention, as Fig. 8 institute
Show.The speech recognition equipment of the present embodiment can include original audio data acquiring unit 81, such as Fig. 5～figure
Noise elimination apparatus 82 and voice recognition unit 83 that embodiment corresponding to 7 either figure is provided.Its
In, original audio data acquiring unit 81, for obtaining pending original audio data；Speech recognition list
Unit 83, for carrying out voice recognition processing to effective voice data.
In the present invention, the detailed description of noise elimination apparatus 82, may refer to corresponding to Fig. 5～Fig. 7
Related content in embodiment, here is omitted.
It should be noted that the speech recognition equipment that provided of the present embodiment can be partly or entirely position
In the application of local terminal, or can also be the plug-in unit or software development being located locally in the application of terminal
The functional units such as kit (Software Development Kit, SDK), or can also be positioned at
Process engine in the server of network side, or can also be the distributed system positioned at network side, this
Embodiment is not particularly limited to this.
It is understood that the application can be mounted in the local program in terminal (nativeApp),
Or can also be a web page program (webApp) of browser in terminal, the present embodiment to this not
It is particularly limited.
It should be noted that method in the corresponding embodiment of Fig. 4, the language that all can be provided by the present embodiment
Sound identifying device is realized.Describe the related content that may refer in the corresponding embodiment of Fig. 4 in detail, herein
Repeat no more.
In the present embodiment, pending original audio data is obtained by original audio data acquiring unit, is entered
And specific vocal print parameter is based on by voice print matching unit, to acquired described pending original audio data
Carry out voice print matching so that effectively voice data acquiring unit can be according to the vocal print of the voice print matching
Join result, effective voice data is obtained from the pending original audio data, and by speech recognition list
Unit voice recognition processing is carried out to effective voice data, due to effective voice data be according to specific sound
Line parameter, the voice data extracted in pending original audio data, this portion of audio data can
To be considered the voice signal of the user of using terminal, therefore, no longer include in effective voice data and make an uproar
Acoustical signal, so as to effectively increase sound quality.
Further, voice recognition processing is carried out to effective voice data, the recognition result for being obtained,
Accuracy rate is higher.
In addition, the technical scheme provided using the present invention, it is only necessary to a voice collection device, Neng Gouyou
Effect reduces cost.
Those skilled in the art can be understood that, for convenience and simplicity of description, above-mentioned retouch
The specific work process of the system that states, equipment and unit, may be referred to the correspondence in preceding method embodiment
Process, will not be described here.
In several embodiments provided by the present invention, it should be understood that disclosed system, equipment and
Method, can be realized by another way.For example, apparatus embodiments described above are only to show
Meaning property, for example, the division of the unit, only a kind of division of logic function, can when actually realizing
To there is other dividing mode, for example multiple units or component can in conjunction with or be desirably integrated into another
System, or some features can ignore, or do not execute.Another, shown or discussed each other
Coupling direct-coupling or communication connection can be the INDIRECT COUPLING of equipment or unit by some interfaces
Or communication connection, can be electrical, mechanical or other forms.
The unit that illustrates as separating component can be or may not be physically separate, make
For the part that unit shows can be or may not be physical location, you can so that a place is located at,
Or can also be distributed on multiple NEs.Can select according to the actual needs part therein or
The whole units of person are realizing the purpose of this embodiment scheme.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit,
Can also be that unit is individually physically present, it is also possible to which two or more units are integrated in a list
In unit.Above-mentioned integrated unit both can be realized in the form of hardware, it would however also be possible to employ hardware adds software
The form of functional unit is realized.
The above-mentioned integrated unit that is realized in the form of SFU software functional unit, can be stored in a computer
In read/write memory medium.Above-mentioned SFU software functional unit is stored in a storage medium, including some fingers
Order is used so that a computer equipment (can be personal computer, server, or network equipment etc.)
Or processor (processor) executes the part steps of each embodiment methods described of the present invention.And it is aforementioned
Storage medium include：USB flash disk, portable hard drive, read-only storage (Read-Only Memory, ROM),
Random access memory (Random Access Memory, RAM), magnetic disc or CD etc. are various
Can be with the medium of store program codes.
Finally it should be noted that：Above example only in order to technical scheme to be described, rather than to which
Limit；Although being described in detail to the present invention with reference to the foregoing embodiments, the ordinary skill of this area
Personnel should be understood：Which still can be modified to the technical scheme described in foregoing embodiments, or
Person carries out equivalent to which part technical characteristic；And these modifications or replacement, do not make corresponding skill
The essence of art scheme departs from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (16)
1. a kind of noise cancellation method, it is characterised in that include：
Based on specific vocal print parameter, voice print matching is carried out to acquired pending original audio data；
According to the voice print matching result of the voice print matching, obtain from the pending original audio data
Effective voice data.
2. noise cancellation method according to claim 1, it is characterised in that the specific vocal print ginseng
Number is the vocal print parameter of targeted customer, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is obtained from the pending original audio data, as described
Effective voice data.
3. noise cancellation method according to claim 2, it is characterised in that described based on specific
Vocal print parameter, before carrying out voice print matching to acquired pending original audio data, methods described is also
Including：
Obtain the voice signal of the targeted customer；
Based on the voice signal of the acquired targeted customer, the vocal print parameter of the targeted customer is obtained.
4. noise cancellation method according to claim 1, it is characterised in that the specific vocal print ginseng
Number is the vocal print parameter of the noise signal of target environment, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is removed from the pending original audio data, as described
Effective voice data.
5. noise cancellation method according to claim 4, it is characterised in that described based on specific
Vocal print parameter, before carrying out voice print matching to acquired pending original audio data, methods described is also
Including：
Obtain the noise signal of the target environment；
Based on the noise signal of the acquired target environment, the vocal print parameter of the noise signal is obtained.
6. a kind of noise elimination apparatus, it is characterised in that include：
Voice print matching unit, for being based on specific vocal print parameter, to acquired pending original audio number
According to carrying out voice print matching；
Effective voice data acquiring unit, for the voice print matching result according to the voice print matching, from institute
State and in pending original audio data, obtain effective voice data.
7. noise elimination apparatus according to claim 6, it is characterised in that the specific vocal print ginseng
Number is the vocal print parameter of targeted customer, and
Effective voice data acquiring unit is used for obtaining vocal print from the pending original audio data
The voice data that the match is successful, used as effective voice data.
8. noise elimination apparatus according to claim 7, it is characterised in that the noise eliminates dress
Putting also includes：
Voice signal acquiring unit, for obtaining the voice signal of the targeted customer；
First vocal print gain of parameter unit, for the voice signal based on the acquired targeted customer,
Obtain the vocal print parameter of the targeted customer.
9. noise elimination apparatus according to claim 6, it is characterised in that the specific vocal print ginseng
Number is the vocal print parameter of the noise signal of target environment, and
Effective voice data acquiring unit is used for removing vocal print from the pending original audio data
The voice data that the match is successful, used as effective voice data.
10. noise elimination apparatus according to claim 9, it is characterised in that the noise is eliminated
Device also includes：
Noise signal acquiring unit, for obtaining the noise signal of the target environment；
Second vocal print gain of parameter unit, for the noise signal based on the acquired target environment,
Obtain the vocal print parameter of the noise signal.
11. a kind of audio recognition methods, it is characterised in that include：
Obtain pending original audio data；
Based on specific vocal print parameter, voice print matching is carried out to acquired described pending original audio data；
According to the voice print matching result of the voice print matching, obtain from the pending original audio data
Effective voice data；
Voice recognition processing is carried out to effective voice data.
12. audio recognition methods according to claim 11, it is characterised in that the specific vocal print
Parameter is the vocal print parameter of targeted customer, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is obtained from the pending original audio data, as described
Effective voice data.
13. audio recognition methods according to claim 12, it is characterised in that described based on spy
Determine vocal print parameter, before carrying out voice print matching to acquired described pending original audio data, described
Method also includes：
Obtain the voice signal of the targeted customer；
Based on the voice signal of the acquired targeted customer, the vocal print parameter of the targeted customer is obtained.
14. audio recognition methods according to claim 11, it is characterised in that the specific vocal print
Parameter is the vocal print parameter of the noise signal of target environment, and
The voice print matching result according to the voice print matching, from the pending original audio data
Effective voice data is obtained, including：
The successful voice data of voice print matching is removed from the pending original audio data, as described
Effective voice data.
15. audio recognition methods according to claim 14, it is characterised in that described based on spy
Determine vocal print parameter, before carrying out voice print matching to acquired described pending original audio data, described
Method also includes：
Obtain the noise signal of the target environment；
Based on the noise signal of the acquired target environment, the vocal print parameter of the noise signal is obtained.
16. a kind of speech recognition equipments, it is characterised in that include：
Original audio data acquiring unit, for obtaining pending original audio data；
Noise elimination apparatus as described in any claim in claim 6 to 10；
Voice recognition unit, for carrying out voice recognition processing to effective voice data.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510524909.1A CN106486130B (en) | 2015-08-25 | 2015-08-25 | Noise elimination and voice recognition method and device |
| PCT/CN2015/095364 WO2017031846A1 (en) | 2015-08-25 | 2015-11-24 | Noise elimination and voice recognition method, apparatus and device, and non-volatile computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510524909.1A CN106486130B (en) | 2015-08-25 | 2015-08-25 | Noise elimination and voice recognition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106486130A true CN106486130A (en) | 2017-03-08 |
| CN106486130B CN106486130B (en) | 2020-03-31 |
Family
ID=58099552
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510524909.1A Active CN106486130B (en) | 2015-08-25 | 2015-08-25 | Noise elimination and voice recognition method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106486130B (en) |
| WO (1) | WO2017031846A1 (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107172256A (en) * | 2017-07-27 | 2017-09-15 | 广东欧珀移动通信有限公司 | Earphone call self-adapting regulation method, device, mobile terminal and storage medium |
| CN107293293A (en) * | 2017-05-22 | 2017-10-24 | 深圳市搜果科技发展有限公司 | A kind of voice instruction recognition method, system and robot |
| CN107945815A (en) * | 2017-11-27 | 2018-04-20 | 歌尔科技有限公司 | Voice signal noise-reduction method and equipment |
| CN108062949A (en) * | 2017-12-11 | 2018-05-22 | 广州朗国电子科技有限公司 | The method and device of voice control treadmill |
| CN108171118A (en) * | 2017-12-05 | 2018-06-15 | 东软集团股份有限公司 | Signal of blinking data processing method, device, readable storage medium storing program for executing and electronic equipment |
| CN108305637A (en) * | 2018-01-23 | 2018-07-20 | 广东欧珀移动通信有限公司 | Earphone method of speech processing, terminal device and storage medium |
| CN109166575A (en) * | 2018-07-27 | 2019-01-08 | 百度在线网络技术（北京）有限公司 | Exchange method, device, smart machine and the storage medium of smart machine |
| CN109360580A (en) * | 2018-12-11 | 2019-02-19 | 珠海市微半导体有限公司 | A kind of iterated denoising device and clean robot based on speech recognition |
| CN109410938A (en) * | 2018-11-28 | 2019-03-01 | 途客电力科技（天津）有限公司 | Control method for vehicle, device and car-mounted terminal |
| CN109493870A (en) * | 2018-11-28 | 2019-03-19 | 途客电力科技（天津）有限公司 | Charging pile identity identifying method, device and electronic equipment |
| CN111696565A (en) * | 2020-06-05 | 2020-09-22 | 北京搜狗科技发展有限公司 | Voice processing method, apparatus and medium |
| CN112992153A (en) * | 2021-04-27 | 2021-06-18 | 太平金融科技服务（上海）有限公司 | Audio processing method, voiceprint recognition device and computer equipment |
| CN113347519A (en) * | 2020-02-18 | 2021-09-03 | 宏碁股份有限公司 | Method for eliminating specific object voice and ear-wearing type sound signal device using same |
| CN113516994A (en) * | 2021-04-07 | 2021-10-19 | 北京大学深圳研究院 | Real-time voice recognition method, device, equipment and medium |
| CN115440198A (en) * | 2022-11-08 | 2022-12-06 | 南方电网数字电网研究院有限公司 | Method and apparatus for converting mixed audio signal, computer device and storage medium |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111161719B (en) * | 2018-11-08 | 2023-01-20 | 联想新视界（北京）科技有限公司 | AR (augmented reality) glasses operated by voice and method for operating AR glasses by voice |
| CN110060689A (en) * | 2019-04-10 | 2019-07-26 | 南京启诺信息技术有限公司 | A kind of intelligent wearable device and its application method with the identification of feature voice |
| CN110265038B (en) * | 2019-06-28 | 2021-10-22 | 联想(北京)有限公司 | Processing method and electronic equipment |
| CN110708625A (en) * | 2019-09-25 | 2020-01-17 | 华东师范大学 | Intelligent terminal-based environment sound suppression and enhancement adjustable earphone system and method |
| CN111883159A (en) * | 2020-08-05 | 2020-11-03 | 龙马智芯（珠海横琴）科技有限公司 | Voice processing method and device |
| CN111951802A (en) * | 2020-08-10 | 2020-11-17 | 山东金东数字创意股份有限公司 | Visual expression system and method based on AI voice recognition |
| CN112464021A (en) * | 2020-11-25 | 2021-03-09 | 上海眼控科技股份有限公司 | Anti-law audio and video determination method, device, equipment and storage medium |
| CN114299981A (en) * | 2021-12-29 | 2022-04-08 | 中国电信股份有限公司 | Audio processing method, device, storage medium and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060149547A1 (en) * | 2005-01-06 | 2006-07-06 | Fuji Photo Film Co., Ltd. | Recording apparatus and voice recorder program |
| US20070263846A1 (en) * | 2006-04-03 | 2007-11-15 | Fratti Roger A | Voice-identification-based signal processing for multiple-talker applications |
| CN102781075A (en) * | 2011-05-12 | 2012-11-14 | 中兴通讯股份有限公司 | Method for reducing communication power consumption of mobile terminal and mobile terminal |
| CN103594092A (en) * | 2013-11-25 | 2014-02-19 | 广东欧珀移动通信有限公司 | Single microphone voice noise reduction method and device |
| CN103971696A (en) * | 2013-01-30 | 2014-08-06 | 华为终端有限公司 | Method, device and terminal equipment for processing voice |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005181391A (en) * | 2003-12-16 | 2005-07-07 | Sony Corp | Device and method for speech processing |
| CN102694891A (en) * | 2011-03-21 | 2012-09-26 | 鸿富锦精密工业（深圳）有限公司 | System and method for removing conversation noises |
| CN103165131A (en) * | 2011-12-17 | 2013-06-19 | 富泰华工业（深圳）有限公司 | Voice processing system and voice processing method |
-
2015
- 2015-08-25 CN CN201510524909.1A patent/CN106486130B/en active Active
- 2015-11-24 WO PCT/CN2015/095364 patent/WO2017031846A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060149547A1 (en) * | 2005-01-06 | 2006-07-06 | Fuji Photo Film Co., Ltd. | Recording apparatus and voice recorder program |
| US20070263846A1 (en) * | 2006-04-03 | 2007-11-15 | Fratti Roger A | Voice-identification-based signal processing for multiple-talker applications |
| CN102781075A (en) * | 2011-05-12 | 2012-11-14 | 中兴通讯股份有限公司 | Method for reducing communication power consumption of mobile terminal and mobile terminal |
| CN103971696A (en) * | 2013-01-30 | 2014-08-06 | 华为终端有限公司 | Method, device and terminal equipment for processing voice |
| CN103594092A (en) * | 2013-11-25 | 2014-02-19 | 广东欧珀移动通信有限公司 | Single microphone voice noise reduction method and device |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107293293A (en) * | 2017-05-22 | 2017-10-24 | 深圳市搜果科技发展有限公司 | A kind of voice instruction recognition method, system and robot |
| CN107172256A (en) * | 2017-07-27 | 2017-09-15 | 广东欧珀移动通信有限公司 | Earphone call self-adapting regulation method, device, mobile terminal and storage medium |
| CN107945815A (en) * | 2017-11-27 | 2018-04-20 | 歌尔科技有限公司 | Voice signal noise-reduction method and equipment |
| CN108171118A (en) * | 2017-12-05 | 2018-06-15 | 东软集团股份有限公司 | Signal of blinking data processing method, device, readable storage medium storing program for executing and electronic equipment |
| CN108062949A (en) * | 2017-12-11 | 2018-05-22 | 广州朗国电子科技有限公司 | The method and device of voice control treadmill |
| CN108305637A (en) * | 2018-01-23 | 2018-07-20 | 广东欧珀移动通信有限公司 | Earphone method of speech processing, terminal device and storage medium |
| CN108305637B (en) * | 2018-01-23 | 2021-04-06 | Oppo广东移动通信有限公司 | Earphone voice processing method, terminal equipment and storage medium |
| CN109166575A (en) * | 2018-07-27 | 2019-01-08 | 百度在线网络技术（北京）有限公司 | Exchange method, device, smart machine and the storage medium of smart machine |
| CN109410938A (en) * | 2018-11-28 | 2019-03-01 | 途客电力科技（天津）有限公司 | Control method for vehicle, device and car-mounted terminal |
| CN109493870A (en) * | 2018-11-28 | 2019-03-19 | 途客电力科技（天津）有限公司 | Charging pile identity identifying method, device and electronic equipment |
| CN109360580A (en) * | 2018-12-11 | 2019-02-19 | 珠海市微半导体有限公司 | A kind of iterated denoising device and clean robot based on speech recognition |
| CN109360580B (en) * | 2018-12-11 | 2022-01-04 | 珠海一微半导体股份有限公司 | Iteration denoising device and cleaning robot based on voice recognition |
| CN113347519A (en) * | 2020-02-18 | 2021-09-03 | 宏碁股份有限公司 | Method for eliminating specific object voice and ear-wearing type sound signal device using same |
| CN113347519B (en) * | 2020-02-18 | 2022-06-17 | 宏碁股份有限公司 | Method for eliminating specific object voice and ear-wearing type sound signal device using same |
| CN111696565A (en) * | 2020-06-05 | 2020-09-22 | 北京搜狗科技发展有限公司 | Voice processing method, apparatus and medium |
| CN111696565B (en) * | 2020-06-05 | 2023-10-10 | 北京搜狗科技发展有限公司 | Voice processing method, device and medium |
| CN113516994A (en) * | 2021-04-07 | 2021-10-19 | 北京大学深圳研究院 | Real-time voice recognition method, device, equipment and medium |
| CN113516994B (en) * | 2021-04-07 | 2022-04-26 | 北京大学深圳研究院 | Real-time voice recognition method, device, equipment and medium |
| CN112992153B (en) * | 2021-04-27 | 2021-08-17 | 太平金融科技服务（上海）有限公司 | Audio processing method, voiceprint recognition device and computer equipment |
| CN112992153A (en) * | 2021-04-27 | 2021-06-18 | 太平金融科技服务（上海）有限公司 | Audio processing method, voiceprint recognition device and computer equipment |
| CN115440198A (en) * | 2022-11-08 | 2022-12-06 | 南方电网数字电网研究院有限公司 | Method and apparatus for converting mixed audio signal, computer device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106486130B (en) | 2020-03-31 |
| WO2017031846A1 (en) | 2017-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106486130A (en) | Noise elimination, audio recognition method and device | |
| CN110085251B (en) | Human voice extraction method, human voice extraction device and related products | |
| CN105023573B (en) | It is detected using speech syllable/vowel/phone boundary of auditory attention clue | |
| CN110648692B (en) | Voice endpoint detection method and system | |
| CN106504768B (en) | Phone testing audio frequency classification method and device based on artificial intelligence | |
| CN107274916B (en) | Method and device for operating audio/video file based on voiceprint information | |
| CN109817236A (en) | Audio defeat method, apparatus, electronic equipment and storage medium based on scene | |
| CN109036436A (en) | A kind of voice print database method for building up, method for recognizing sound-groove, apparatus and system | |
| CN108962260A (en) | A kind of more human lives enable audio recognition method, system and storage medium | |
| CN108564963A (en) | Method and apparatus for enhancing voice | |
| CN109657739B (en) | Handwritten letter identification method based on high-frequency sound wave short-time Fourier transform | |
| CN103971696A (en) | Method, device and terminal equipment for processing voice | |
| CN108198569A (en) | A kind of audio-frequency processing method, device, equipment and readable storage medium storing program for executing | |
| CN109394258A (en) | A kind of classification method, device and the terminal device of lung's breath sound | |
| CN108986798B (en) | Processing method, device and the equipment of voice data | |
| CN104142831B (en) | Application program searching method and device | |
| WO2019242402A1 (en) | Speech recognition model generation method and apparatus, and storage medium and electronic device | |
| CN110473552A (en) | Speech recognition authentication method and system | |
| CN109192192A (en) | A kind of Language Identification, device, translator, medium and equipment | |
| CN106228047B (en) | A kind of application icon processing method and terminal device | |
| CN111179909B (en) | Multi-microphone far-field voice awakening method and system | |
| CN105913842A (en) | Method for waking up mobile phone by custom voice | |
| CN110334497A (en) | Switching method and wearable electronic equipment, the storage medium of display interface | |
| CN105590629B (en) | A kind of method and device of speech processes | |
| CN113327586A (en) | Voice recognition method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |