CN104378774A - Voice quality processing method and device - Google Patents
Voice quality processing method and device Download PDFInfo
- Publication number
- CN104378774A CN104378774A CN201310355855.1A CN201310355855A CN104378774A CN 104378774 A CN104378774 A CN 104378774A CN 201310355855 A CN201310355855 A CN 201310355855A CN 104378774 A CN104378774 A CN 104378774A
- Authority
- CN
- China
- Prior art keywords
- voice signal
- parameter
- noise
- described voice
- carry out
- 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
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal 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
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/01—Assessment or evaluation of speech recognition systems
Abstract
The invention discloses a voice quality processing method and device and relates to the technical field of communication. The method comprises the steps of carrying out noise detection on voice signals to be processed to obtain the current noise environment of the voice signals; obtaining adaptive voice processing parameters according to the current noise environment and a voice quality mode selected by a user; carrying out voice processing on the voice signals by utilizing the voice processing parameters so that the voice quality of the processed voice signals can be adaptive to the current noise environment and the voice quality mode selected by the user. According to the method, adaptive voice processing parameters can be obtained through the current noise environment and the voice quality mode selected by the user, different voice quality requests under different noise environments can be met, and therefore individual needs of different users under different noise environments can be met.
Description
Technical field
The present invention relates to communication technical field, the method for particularly voice call quality process under a kind of varying environment and device.
Background technology
Current, universal along with intelligent terminal, many terminals all have employed algorithm with de-noising function or chip.And in current terminal, in a handheld mode, even if user is in different environment, terminal all adopts a set of voice software algorithm parameter to eliminate noise, echo etc.
Along with terminal scene that is more and more intelligent, that be suitable for gets more and more, adopt the limitation of a set of voice software algorithm parameter just increasing, and different user subjective sense of hearing requirement under various circumstances can not be met.In different environments, the people of different identity also has different requirements to voice quality, in public places or meeting room internal reference add videoconference, different people has different demands, someone needs high fidelity to distinguish the people of speech, somebody is more responsive to noise size, and somebody is to insensitive for noise, but strong to the size requirements of sound; The people of some specific place or specific identity, mobile phone communication both sides wish that both talk as whispered, prevent from being heard by other people ... this processes the fidelity, noise, echo etc. of voice accordingly with regard to needing the parameter adjustment disparate modules of hardware being carried out to otherness, thus meets the individual demand of user under different scene.
Summary of the invention
The object of the present invention is to provide a kind of method and device of voice quality process, better can solve and a set of voice software algorithm parameter can only be adopted can not to meet the confinement problems of user to the individual demand under varying environment.
According to an aspect of the present invention, provide a kind of method of voice quality process, comprising:
By carrying out noise measuring to pending voice signal, obtain the current noise environment of described voice signal;
According to the voice quality pattern that described current noise environment and user are selected, obtain adaptive speech parameters;
Utilize described speech parameters to carry out speech processes to described voice signal, the voice quality pattern that the voice quality of the voice signal after process and current noise environment and user are selected adapts.
Preferably, described by carrying out noise measuring to pending voice signal, the current noise environment obtaining described voice signal comprises:
By carrying out analog-to-digital conversion to the analog voice signal of microphone pickup, obtain audio digital signals;
By carrying out noise measuring to described audio digital signals, obtain current noise environmental parameter;
Utilize preset noise circumstance parameter and the relation table of noise circumstance, obtain (transmit leg) current noise environment corresponding with current noise environmental parameter.
Preferably, described by carrying out noise measuring to pending voice signal, the current noise environment obtaining described voice signal also comprises:
Carry out decoding process by the audio digital signals of the coding received antenna, obtain the audio digital signals of decoding;
By carrying out noise measuring to the audio digital signals of described decoding, obtain the noise circumstance parameter of current the other side;
Utilize preset noise circumstance parameter and the relation table of noise circumstance, obtain (recipient) current noise environment corresponding with the noise circumstance parameter of current the other side.
Preferably, the described voice quality pattern selected according to described current noise environment and user, the step obtaining adaptive speech parameters comprises:
Search the parameter that the voice quality pattern selected with current noise environment and user adapts in the multiple parameters grouping of preserving from parameter grouping module to divide into groups;
Parameter configuration in the parameter found being divided into groups is elected to be described speech parameters.
Preferably, the described speech parameters that utilizes comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Balance parameters is utilized to carry out equilibrium treatment to the voice signal carrying out denoising Processing, to adjust the distortion factor of described voice signal.
Preferably, the described speech parameters that utilizes also comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal.
Preferably, the described speech parameters that utilizes also comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Filtering parameter is utilized to carry out filtering process, with the noise beyond voice signal midband described in filtering to the voice signal carrying out denoising Processing.
Preferably, the described speech parameters that utilizes also comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Filtering parameter is utilized to carry out filtering process, with the noise beyond voice signal midband described in filtering to the voice signal carrying out denoising Processing;
The echo parameter that disappears is utilized to disappear echo process to the voice signal carrying out filtering process, to eliminate the echo in described voice signal;
The voice signal of gain parameter to the echo process that disappears is utilized to carry out gain amplification disposal, to amplify the gain of described voice signal;
Balance parameters is utilized to carry out equilibrium treatment to the voice signal carrying out gain amplification, to adjust the distortion factor of described voice signal.
According to a further aspect in the invention, provide a kind of device of voice quality process, comprising:
Noise circumstance module, for by carrying out noise measuring to pending voice signal, obtains the current noise environment of described voice signal;
Main processor modules, for the voice quality pattern selected according to described current noise environment and user, obtains adaptive speech parameters;
Speech processing module, for utilizing described speech parameters to carry out speech processes to described voice signal, the voice quality pattern that the voice quality of the voice signal after process and current noise environment and user are selected adapts.
Preferably, described speech processing module comprises further:
De-noising submodule, for utilizing de-noising parameter to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Balanced submodule, for utilizing balance parameters to carry out equilibrium treatment to described voice signal, to adjust the distortion factor of described voice signal;
Filtering submodule, for utilizing filtering parameter to carry out filtering process to described voice signal, to eliminate the noise in described voice signal;
Disappear echo submodule, to disappear echo process for utilizing the echo parameter that disappears, to eliminate the echo in described voice signal to described voice signal;
Gain submodule, for utilizing gain parameter to carry out gain amplification disposal to described voice signal, to amplify the gain of described voice signal.
Compared with prior art, beneficial effect of the present invention is: the present invention can according to user's different demands in certain circumstances, voice quality pattern required for selection, and by different noise circumstance and different user to the demand of voice quality, obtain the method for adaptive speech parameters, realize different user in different environments to the individual demand of voice quality.
Accompanying drawing explanation
Fig. 1 is the method flow diagram of the voice quality process that the embodiment of the present invention provides;
Fig. 2 is the structure drawing of device of the voice quality process that the embodiment of the present invention provides;
Fig. 3 is the co-operating transmit leg software flow block diagram of intermodule of the voice quality process that the embodiment of the present invention provides;
Fig. 4 is the co-operating recipient's software flow block diagram of intermodule of the voice quality process that the embodiment of the present invention provides.
Embodiment
Below in conjunction with accompanying drawing to a preferred embodiment of the present invention will be described in detail, should be appreciated that following illustrated preferred embodiment is only for instruction and explanation of the present invention, is not intended to limit the present invention.
Fig. 1 is the method flow diagram of the voice quality process that the embodiment of the present invention provides, and as shown in Figure 1, step is as follows:
Step S1: by carrying out noise measuring to pending voice signal, obtains the current noise environment of described voice signal.
Step S2: the voice quality pattern selected according to described current noise environment and user, obtains adaptive speech parameters.
Step S3: utilize described speech parameters to carry out speech processes to described voice signal, the voice quality pattern that the voice quality of the voice signal after process and current noise environment and user are selected adapts.
In described step S1, by carrying out analog-to-digital conversion to the analog voice signal of microphone pickup, obtain audio digital signals;
By carrying out noise measuring to described audio digital signals, obtain current noise environmental parameter;
Utilize preset noise circumstance parameter and the relation table of noise circumstance, obtain (transmit leg) current noise environment corresponding with current noise environmental parameter.
Carry out decoding process by the audio digital signals of the coding received antenna, obtain the audio digital signals of decoding;
By carrying out noise measuring to the audio digital signals of described decoding, obtain the noise circumstance parameter of current the other side;
Utilize preset noise circumstance parameter and the relation table of noise circumstance, obtain (recipient) current noise environment corresponding with the noise circumstance parameter of current the other side.
In described step S2, search the parameter that the voice quality pattern selected with current noise environment and user adapts in the multiple parameters grouping of preserving from parameter grouping module and divide into groups;
Parameter configuration in the parameter found being divided into groups is elected to be described speech parameters.
In described step S3, de-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal.
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Balance parameters is utilized to carry out equilibrium treatment to the voice signal carrying out denoising Processing, to adjust the distortion factor of described voice signal.
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Filtering parameter is utilized to carry out filtering process, with the noise beyond voice signal midband described in filtering to the voice signal carrying out denoising Processing.
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Filtering parameter is utilized to carry out filtering process, with the noise beyond voice signal midband described in filtering to the voice signal carrying out denoising Processing;
The echo parameter that disappears is utilized to disappear echo process to the voice signal carrying out filtering process, to eliminate the echo in described voice signal;
The voice signal of gain parameter to the echo process that disappears is utilized to carry out gain amplification disposal, to amplify the gain of described voice signal;
Balance parameters is utilized to carry out equilibrium treatment to the voice signal carrying out gain amplification, to adjust the distortion factor of described voice signal.
The embodiment of the present invention 1 provides a kind of solution and realize a certain voice quality under different noise circumstance---the method for high-fidelity pattern.Fig. 2 is the structure drawing of device of the voice quality process that the embodiment of the present invention provides, and as shown in Figure 2, comprising: upper application module, Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module, speech processing module, parameter grouping module, main processor modules.
Speech parameters module in described main processor modules is used for the voice quality pattern selected according to described current noise environment and user, obtains adaptive speech parameters.
Voice-quality module in described speech processing module is used for utilizing described speech parameters to carry out speech processes to described voice signal, and the voice quality pattern that the voice quality of the voice signal after process and current noise environment and user are selected adapts.Wherein, the de-noising submodule of described voice-quality module is used for utilizing de-noising parameter to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal.The balanced submodule of described voice-quality module is used for utilizing balance parameters to carry out equilibrium treatment to described voice signal, to adjust the distortion factor of described voice signal.The filtering submodule of described voice-quality module is used for utilizing filtering parameter to carry out filtering process to described voice signal, to eliminate the noise in described voice signal.The echo submodule that disappears of described voice-quality module to disappear echo process to described voice signal for utilizing the echo parameter that disappears, to eliminate the echo in described voice signal.The gain submodule of described voice-quality module is used for utilizing gain parameter to carry out gain amplification disposal to described voice signal, to amplify the gain of described voice signal.Wherein speech processing module is not limited to concrete realization, can use the Processing Algorithm of software, also can use digital signal processor (the DSP:Digital Signal Processor) chip of hardware.
Described Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module, speech processing module and main processor modules achieve the function of noise circumstance module, for carrying out noise measuring by described Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module and speech processing module to pending voice signal, and in described main processor modules, obtain the current noise environment of described voice signal.
The operation principle of described device is as follows, is described respectively to transmit leg and recipient:
Step 201: sending direction path, Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module by microphone pickup to analog voice signal carry out analog-to-digital conversion, convert audio digital signals to.
Step 202: audio digital signals is sent into speech processing module by digital interface, first speech processing module carries out noise measuring to described audio digital signals, obtain current noise circumstance parameter, and sent into main processor modules to determine the noise circumstance residing for active user.
Step 203: main processor modules is after the parameter of high-fidelity pattern obtaining user's needs that upper application module issues, using described high-fidelity mode parameter and current noise circumstance parameter as correlated judgment condition, by the parameters grouping in relevant interface query argument grouping module.
Step 204: be met the parameter configuration of both conditions from parameter grouping after, reads interface by relevant, reads described parameter configuration, and by relevant interface of writing by described parameter configuration write speech processing module.
Step 205: speech processing module carries out respective handling according to the parameter configuration of write to noisy speech signals, under high-fidelity pattern, by the frequency spectrum parameter of write, regulate equalization filter and relevant high-pass and low-pass filter, frequency spectrum adjustment is carried out to voice signal, by regulating the voice signal frequency range of distortion, it is made to reach undistorted effect.
Step 206: by noise, the echo parameter of write, call relevant de-noising, disappear echo submodule, carries out de-noising to current noise and to disappear echo process, finally reaches the effect that user needs.
Step 207: after having processed, sends the audio digital signals after process into main processor modules and carries out subsequent treatment.
Step 208: receive direction path, decoded audio digital signals is sent into speech processing module by digital interface by main processor modules.
Step 209: speech processing module carries out noise measuring to decoded audio digital signals, obtains current noise circumstance parameter, and is sent into main processor modules to determine the noise circumstance residing for current the other side.
Step 210: main processor modules is after the parameter of high-fidelity pattern obtaining user's needs that upper application module issues, using described high-fidelity mode parameter and current noise circumstance parameter as correlated judgment condition, by the parameters grouping in relevant interface inquiry packets module.
Step 211: be met the parameter configuration of both conditions from parameter grouping after, reads interface by relevant, reads described parameter configuration.And by relevant interface of writing by described configuration parameter write speech processing module.
Step 212: speech processing module carries out respective handling according to write to noisy speech signals, under high-fidelity pattern, by the frequency spectrum parameter of write, regulate equalization filter and relevant high-pass and low-pass filter, frequency spectrum adjustment is carried out to decoded voice signal, by regulating the voice signal frequency range of distortion, it is made to reach undistorted effect.
Step 213: by noise, the echo parameter of write, call relevant de-noising, disappear echo submodule, carries out de-noising to current noise and to disappear echo process, finally reaches the effect that user needs.
Step 214: after having processed, sends into Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module by the audio digital signals after process by digital interface, converts audio digital signals to analog voice signal, sends into the devices such as loud speaker and plays back.
In above-mentioned general principle, parameter grouping module is divided into groups based on the voice quality pattern of user's request in the current residing environment of user and upper application module, so in embodiments, the environment residing for active user is divided into according to noise intensity: quiet environment, slight noise circumstance, moderate noise environment, strong noise environment etc.The noise intensity of wherein arranging each environment is within the scope of one, such as, in typical moderate noise environment (as working environment), noise intensity is certain, so increases the noise intensity in this environment, after acquiring a certain degree, becomes strong noise environment.So when the noise intensity of the current residing environment of user is in the noise range of some noise circumstances, we just process with the representative value of this noise circumstance.Certainly the classification for noise situations is not limited to concrete above-mentioned form, it can have other classification form, the maximum arranged in noise range as often kind of noise circumstance carries out noise processed etc., and concrete way of realization can realize according to the demand of user and concrete design.
Under high-fidelity pattern, based on the difference of noise circumstance, the parameter of high-fidelity pattern is divided into multiple parameter to divide into groups according to the difference of noise intensity or the difference of noise type, and the parameter of high-fidelity pattern is regulated according to different noise circumstances, re-writes parameters grouping.Wherein the parameter of the high-fidelity pattern of in advance write carries out regulating based on the calibration curve of the sound of people, and under different noise circumstances, based on the speech signal spec-trum of reality, the effect of adjustment can be different, thus the parameter configuration obtained is also different.Such as under comparatively quiet environment, noise intensity is less, do not need to use high-pass and low-pass filter to carry out stronger filtering to voice signal, only need de-noising submodule to carry out the process just energy stress release treatment of some strength, and repair by regulating frequency spectrum to regulate the equalization filter in submodule to carry out adjustment to the voice signal of distortion; If residing noise circumstance is more severe, in order to the degree making the voice signal of needs reach high-fidelity, by equalization filter, stronger filtering is not carried out to noisy speech signals, the filtering can only carrying out to a certain degree to noisy speech signals by de-noising submodule, slight noisy speech signals can be mixed with in the signal of now high-fidelity, namely be that cost exchanges the undistorted of voice signal for noise, thus reach the effect of high-fidelity.
The embodiment of the present invention 1 is just described with the high-fidelity pattern in upper application module, in the process of call, if user selects comfort level pattern, step is similar with the embodiment of the present invention 1, namely, under different noise circumstances, main processor modules can read the parameter configuration in parameter grouping module under comfort level pattern in corresponding noise intensity.By this parameter configuration, write speech processing module, speech processing module can carry out adjustment reparation to the distorted speech signal under current noise environment, and by de-noising submodule filtering noise wherein, by the echo submodule elimination echo wherein that disappears.After finally overall process being carried out to voice signal, ensure voice signal hardly distortion with almost there is no noise, make the effect that speech quality reaches subjective pleasant.
If user selects low noise mode, other steps are similar with the embodiment of the present invention 1, difference is based on different noise circumstances, specifically can regulate, be applicable to the parameter configuration of the low noise mode of different noise circumstance respectively to the parameter of low noise mode.Such as when ambient intensity is quiet environment, for the de-noising parameter needing the low noise mode regulated, making it reach disposal ability is not very strong effect; And when ambient intensity is strong noise environment, need the de-noising parameter regulating low noise mode, make the noise processed ability of de-noising submodule stronger, the submodules such as filter can carry out filtering to noisy speech signals.In such cases, exchange the whole eliminations to nonstationary noise for the sacrificial section distortion factor, thus reach the low noise voice quality pattern of user's needs, it is comparatively severe that this pattern is applicable to environment, simultaneously to the insensitive user of audio distortions degree.
If user selects privacy mode, because in such a mode, the speech signal energy of the not Noise of input is less, and noisy speech signals is comparatively strong, therefore needs specifically to process voice signal.Concerning transmit leg, after being amplified by analog gain, carry out analog-to-digital conversion through Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module.Wherein after analog gain amplifies, more weak voice signal is exaggerated, and that amplifies also has noisy speech signals simultaneously.Now send containing the voice signal compared with very noisy voice signal into speech processing module, obtain current noise circumstance parameter, main processor modules is according to the parameter of current noise circumstance parameter and privacy mode, read the parameter configuration under privacy mode in corresponding noise intensity in parameter grouping module, speech processing module carries out filtering according to the parameter configuration read to the noisy speech signals be mixed in voice signal.Because noisy speech signals is stronger, so need first to call the out-of-band noisy speech signals of filter sub-module filtering voice signal, call de-noising submodule and filtering is carried out to the noisy speech signals in frequency band, call the echo voice signal disappeared in echo submodule filtering voice signal simultaneously, eventually pass digital gain to amplify, voice signal less for energy is amplified to suitable yield value, through equalization filter, regulate the frequency spectrum of voice signal, make it reach good fidelity.
Fig. 3 is the co-operating transmit leg software flow block diagram of intermodule of the voice quality process that the embodiment of the present invention provides.As shown in Figure 3, be described in detail transmit leg software flow provided by the invention by specific embodiment 1, step is as follows:
Step 301: user makes a call and asks or accept called request.
Step 302: upper level applications calls the related procedure of bottom, completes relevant initial work, and calls startup hardware pipeline.
Step 303: the parameter of the voice quality pattern that upper level applications can be selected user sends to master processor program, and by described reference record in master processor program.
Step 304: start hardware pipeline, by powering on to hardware, distributing the operations such as clock, realize the initialization operation to software, hardware can normally be worked.
Step 305: the noisy analog voice signal that contains of pickup sends to Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module to carry out analog-to-digital conversion by microphone.
Step 306: after Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module convert audio digital signals to the analog voice signal received, send to speech processing module to carry out speech processes by digital interface.
Step 307: carry out after speech detection through speech processing module, obtains current noise circumstance parameter, and current noise circumstance parameter is sent into master processor program by relevant interface.
Step 308: the relevant parameter in step 303 and step 307 as querying condition, is carried out parameter Querying by group by master processor program in parameter grouping module.
Step 309: after inquiring the parameter configuration satisfied condition, master processor program reads relevant parameter configuration.
Step 310: the relevant parameter read configuration is write speech processing module by relevant interface by master processor program.
Step 311: speech processing module, according to the relevant parameter configuration of write, is carried out corresponding function treatment to the audio digital signals containing noise echo, and the audio digital signals after function treatment is sent to master processor program again.
Step 312: after master processor program carries out the work such as coding to the audio digital signals after noise echo function treatment, sends to subsequent treatment module to carry out subsequent treatment the audio digital signals after coding.
Step 313: subsequent treatment module carries out the process such as agreement, physics to the audio digital signals after coding, is sent the audio digital signals after coding by antenna.
Fig. 4 is the co-operating recipient's software flow block diagram of intermodule of the voice quality process that the embodiment of the present invention provides.As shown in Figure 4, be described in detail recipient's software flow provided by the invention by specific embodiment, step is as follows:
Step 401: user makes a call and asks or accept called request.
Step 402: audio digital signals receives by antenna, and process accordingly, the audio digital signals processed is sent into master processor program, carries out the work such as decoding.
Step 403: the parameter of the voice quality pattern that user selects by upper level applications sends to master processor program, and by described reference record in master processor program.
Step 404: audio digital signals is sent into speech processing module and carried out speech detection by master processor program, obtains corresponding noise circumstance parameter.
Step 405: the corresponding noise circumstance parameter obtained is sent to master processor program by relevant interface.
Step 406: master processor program carries out parameter Querying by group as querying condition according to the relevant parameter in step 404 and step 405 in parameter grouping module.
Step 407: after inquiring the parameter configuration satisfied condition, master processor program reads relevant parameter configuration.
Step 408: the relevant parameter read configuration is write speech processing module by relevant interface by master processor program.
Step 409: speech processing module, according to the relevant parameter configuration of write, is carried out corresponding function treatment to the audio digital signals containing noise echo, and the audio digital signals after function treatment is sent to Data Collection/broadcasting and digital-to-analogue/analog-to-digital conversion module.
Step 410: the audio digital signals after function treatment is carried out digital-to-analogue conversion, converts analog voice signal to, play back.
The embodiment of the present invention 2 provides a kind of extending method on embodiment 1 basis, and the difference that described extending method is compared to embodiment 1 is as follows:
Coordinate the devices such as the acceleration sensor in mobile terminal, can detect when external environment condition is quiet, and user is in non-athletic state.In such cases, by regulating the parameter of related voice quality mode, under the prerequisite not damaging voice quality, control voice processor module by main processor modules and close relevant sub-module, as closed automatic growth control/dynamic range control (AGC:Automatic Generation Control/DRC:Dynamic Range Control) function etc.Like this, in the process of extended telephone conversation, the load of relevant sub-module can be reduced, reduce heating power consumption problem; Also can in such cases, need not all carry out query argument configuration to containing noisy voice signal and read the work of parameter configuration at every turn, one-time detection can be carried out by every 2s or longer time to the noise circumstance parameter containing noisy voice signal, reach the object reducing main processor modules load.This expansion scheme is also not only confined to the application of this class, can also coordinate other devices, realize the function of other classes.
The embodiment of the present invention 3 provides a kind of extending method on embodiment 1 basis, and the difference that described extending method is compared to embodiment 1 is as follows:
The condition that embodiment 1 carries out query argument configuration is that the voice quality pattern selected by user and noise circumstance determine jointly.On the basis of embodiment 1, if user does not carry out the selection of voice quality pattern, mobile terminal can carry out query argument configuration according to the voice quality pattern of noise circumstance and acquiescence.Concrete operation realizes being not limited to above-mentioned several situation, depends on the actual demand of user and the realization of design.
In sum, the present invention has following technique effect: the voice quality pattern selected by user and current noise circumstance obtain the method for adaptive speech parameters, realize user's different demands in certain circumstances, by selecting required voice quality pattern, meet the individual demand of user.
Although above to invention has been detailed description, the present invention is not limited thereto, those skilled in the art of the present technique can carry out various amendment according to principle of the present invention.Therefore, all amendments done according to the principle of the invention, all should be understood to fall into protection scope of the present invention.
Claims (10)
1. a method for voice quality process, is characterized in that,
By carrying out noise measuring to pending voice signal, obtain the current noise environment of described voice signal;
According to the voice quality pattern that described current noise environment and user are selected, obtain adaptive speech parameters;
Utilize described speech parameters to carry out speech processes to described voice signal, the voice quality pattern that the voice quality of the voice signal after process and current noise environment and user are selected adapts.
2. method according to claim 1, is characterized in that, described by carrying out noise measuring to pending voice signal, the current noise environment obtaining described voice signal comprises:
By carrying out analog-to-digital conversion to the analog voice signal of microphone pickup, obtain audio digital signals;
By carrying out noise measuring to described audio digital signals, obtain current noise environmental parameter;
Utilize preset noise circumstance parameter and the relation table of noise circumstance, obtain the current noise environment corresponding with current noise environmental parameter.
3. method according to claim 1 and 2, is characterized in that, described by carrying out noise measuring to pending voice signal, the current noise environment obtaining described voice signal also comprises:
Carry out decoding process by the audio digital signals of the coding received antenna, obtain the audio digital signals of decoding;
By carrying out noise measuring to the audio digital signals of described decoding, obtain the noise circumstance parameter of current the other side;
Utilize preset noise circumstance parameter and the relation table of noise circumstance, obtain the current noise environment corresponding with the noise circumstance parameter of current the other side.
4. method according to claim 1, is characterized in that, the described voice quality pattern selected according to described current noise environment and user, and the step obtaining adaptive speech parameters comprises:
Search the parameter that the voice quality pattern selected with current noise environment and user adapts in the multiple parameters grouping of preserving from parameter grouping module to divide into groups;
Parameter configuration in the parameter found being divided into groups is elected to be described speech parameters.
5. method according to claim 1, is characterized in that, the described speech parameters that utilizes comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal.
6. method according to claim 1, is characterized in that, the described speech parameters that utilizes comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Balance parameters is utilized to carry out equilibrium treatment to the voice signal carrying out denoising Processing, to adjust the distortion factor of described voice signal.
7. method according to claim 1, is characterized in that, the described speech parameters that utilizes comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Filtering parameter is utilized to carry out filtering process, with the noise beyond voice signal midband described in filtering to the voice signal carrying out denoising Processing.
8. method according to claim 1, is characterized in that, the described speech parameters that utilizes comprises the step that described voice signal carries out speech processes:
De-noising parameter is utilized to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Filtering parameter is utilized to carry out filtering process, with the noise beyond voice signal midband described in filtering to the voice signal carrying out denoising Processing;
The echo parameter that disappears is utilized to disappear echo process to the voice signal carrying out filtering process, to eliminate the echo in described voice signal;
The voice signal of gain parameter to the echo process that disappears is utilized to carry out gain amplification disposal, to amplify the gain of described voice signal;
Balance parameters is utilized to carry out equilibrium treatment to the voice signal carrying out gain amplification, to adjust the distortion factor of described voice signal.
9. a device for voice quality process, is characterized in that,
Noise circumstance module, for by carrying out noise measuring to pending voice signal, obtains the current noise environment of described voice signal;
Speech parameters module, for the voice quality pattern selected according to described current noise environment and user, obtains adaptive speech parameters;
Voice-quality module, for utilizing described speech parameters to carry out speech processes to described voice signal, the voice quality pattern that the voice quality of the voice signal after process and current noise environment and user are selected adapts.
10. device according to claim 9, is characterized in that, described voice-quality module comprises further:
De-noising submodule, for utilizing de-noising parameter to carry out denoising Processing to described voice signal, to eliminate the noise in described voice signal;
Balanced submodule, for utilizing balance parameters to carry out equilibrium treatment to described voice signal, to adjust the distortion factor of described voice signal;
Filtering submodule, for utilizing filtering parameter to carry out filtering process to described voice signal, to eliminate the noise in described voice signal;
Disappear echo submodule, to disappear echo process for utilizing the echo parameter that disappears, to eliminate the echo in described voice signal to described voice signal;
Gain submodule, for utilizing gain parameter to carry out gain amplification disposal to described voice signal, to amplify the gain of described voice signal.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310355855.1A CN104378774A (en) | 2013-08-15 | 2013-08-15 | Voice quality processing method and device |
| PCT/CN2013/086095 WO2014161299A1 (en) | 2013-08-15 | 2013-10-28 | Voice quality processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310355855.1A CN104378774A (en) | 2013-08-15 | 2013-08-15 | Voice quality processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104378774A true CN104378774A (en) | 2015-02-25 |
Family
ID=51657489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310355855.1A Pending CN104378774A (en) | 2013-08-15 | 2013-08-15 | Voice quality processing method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104378774A (en) |
| WO (1) | WO2014161299A1 (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016180100A1 (en) * | 2015-05-12 | 2016-11-17 | 腾讯科技（深圳）有限公司 | Method and device for improving audio processing performance |
| CN106328154A (en) * | 2015-06-30 | 2017-01-11 | 芋头科技（杭州）有限公司 | Front-end audio processing system |
| CN108964682A (en) * | 2018-08-02 | 2018-12-07 | 北京车联天下信息技术有限公司 | Radio noise-reduction method, denoising device and electronic equipment |
| CN109243481A (en) * | 2018-09-28 | 2019-01-18 | 出门问问信息科技有限公司 | A kind of method of speech processing and device |
| CN109819129A (en) * | 2018-09-18 | 2019-05-28 | 杭州叙简科技股份有限公司 | A kind of meeting comfort noise mixer system and its method based on sound evaluation |
| CN113015050A (en) * | 2019-12-20 | 2021-06-22 | 瑞昱半导体股份有限公司 | Audio playing device and method with anti-noise mechanism |
| TWI739236B (en) * | 2019-12-13 | 2021-09-11 | 瑞昱半導體股份有限公司 | Audio playback apparatus and method having noise-canceling mechanism |
| CN113808605A (en) * | 2021-09-29 | 2021-12-17 | 睿云联(厦门)网络通讯技术有限公司 | Building intercom system-based voice enhancement method, device and equipment |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9407989B1 (en) | 2015-06-30 | 2016-08-02 | Arthur Woodrow | Closed audio circuit |
| CN113531844B (en) * | 2021-07-06 | 2023-01-13 | 青岛海尔空调器有限总公司 | Control method and system for noise reduction of air conditioner, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1965218A (en) * | 2004-06-04 | 2007-05-16 | 皇家飞利浦电子股份有限公司 | Performance prediction for an interactive speech recognition system |
| US20080153537A1 (en) * | 2006-12-21 | 2008-06-26 | Charbel Khawand | Dynamically learning a user's response via user-preferred audio settings in response to different noise environments |
| US20080159560A1 (en) * | 2006-12-30 | 2008-07-03 | Motorola, Inc. | Method and Noise Suppression Circuit Incorporating a Plurality of Noise Suppression Techniques |
| CN101320563A (en) * | 2007-06-05 | 2008-12-10 | 华为技术有限公司 | Background noise encoding/decoding device, method and communication equipment |
| US20110300806A1 (en) * | 2010-06-04 | 2011-12-08 | Apple Inc. | User-specific noise suppression for voice quality improvements |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177728B (en) * | 2011-12-21 | 2015-07-29 | 中国移动通信集团广西有限公司 | Voice signal denoise processing method and device |
-
2013
- 2013-08-15 CN CN201310355855.1A patent/CN104378774A/en active Pending
- 2013-10-28 WO PCT/CN2013/086095 patent/WO2014161299A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1965218A (en) * | 2004-06-04 | 2007-05-16 | 皇家飞利浦电子股份有限公司 | Performance prediction for an interactive speech recognition system |
| US20080153537A1 (en) * | 2006-12-21 | 2008-06-26 | Charbel Khawand | Dynamically learning a user's response via user-preferred audio settings in response to different noise environments |
| US20080159560A1 (en) * | 2006-12-30 | 2008-07-03 | Motorola, Inc. | Method and Noise Suppression Circuit Incorporating a Plurality of Noise Suppression Techniques |
| CN101320563A (en) * | 2007-06-05 | 2008-12-10 | 华为技术有限公司 | Background noise encoding/decoding device, method and communication equipment |
| US20110300806A1 (en) * | 2010-06-04 | 2011-12-08 | Apple Inc. | User-specific noise suppression for voice quality improvements |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10522164B2 (en) | 2015-05-12 | 2019-12-31 | TENCENT TECHNOLOGY (SHENZHEN) COMPANY LlMITED | Method and device for improving audio processing performance |
| AU2016260156B2 (en) * | 2015-05-12 | 2018-08-02 | Tencent Technology (Shenzhen) Company Limited | Method and device for improving audio processing performance |
| WO2016180100A1 (en) * | 2015-05-12 | 2016-11-17 | 腾讯科技（深圳）有限公司 | Method and device for improving audio processing performance |
| CN106328154A (en) * | 2015-06-30 | 2017-01-11 | 芋头科技（杭州）有限公司 | Front-end audio processing system |
| CN108964682A (en) * | 2018-08-02 | 2018-12-07 | 北京车联天下信息技术有限公司 | Radio noise-reduction method, denoising device and electronic equipment |
| CN109819129A (en) * | 2018-09-18 | 2019-05-28 | 杭州叙简科技股份有限公司 | A kind of meeting comfort noise mixer system and its method based on sound evaluation |
| CN109819129B (en) * | 2018-09-18 | 2020-05-29 | 杭州叙简科技股份有限公司 | Conference comfort noise mixing system and method based on sound evaluation |
| CN109243481A (en) * | 2018-09-28 | 2019-01-18 | 出门问问信息科技有限公司 | A kind of method of speech processing and device |
| TWI739236B (en) * | 2019-12-13 | 2021-09-11 | 瑞昱半導體股份有限公司 | Audio playback apparatus and method having noise-canceling mechanism |
| US11361746B2 (en) | 2019-12-13 | 2022-06-14 | Realtek Semiconductor Corporation | Audio playback apparatus and method having noise-canceling mechanism |
| CN113015050A (en) * | 2019-12-20 | 2021-06-22 | 瑞昱半导体股份有限公司 | Audio playing device and method with anti-noise mechanism |
| CN113015050B (en) * | 2019-12-20 | 2022-11-22 | 瑞昱半导体股份有限公司 | Audio playing device and method with anti-noise mechanism |
| CN113808605A (en) * | 2021-09-29 | 2021-12-17 | 睿云联(厦门)网络通讯技术有限公司 | Building intercom system-based voice enhancement method, device and equipment |
| CN113808605B (en) * | 2021-09-29 | 2023-09-12 | 睿云联(厦门)网络通讯技术有限公司 | Voice enhancement method, device and equipment based on building intercom system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2014161299A1 (en) | 2014-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104378774A (en) | Voice quality processing method and device | |
| US8744091B2 (en) | Intelligibility control using ambient noise detection | |
| CN1627635B (en) | Signal compression/expansion device and mobile communication terminal | |
| CA2766196C (en) | Apparatus, method and computer program for controlling an acoustic signal | |
| EP3058563A1 (en) | Limiting active noise cancellation output | |
| CA2542622A1 (en) | Communication headset with signal processing capability | |
| US10325584B2 (en) | Active noise cancelling device and method of actively cancelling acoustic noise | |
| CN1988737B (en) | System for controlling a transfer function of a hearing aid | |
| CN102158778A (en) | Method, equipment and system for reducing headset noise | |
| CN101478589A (en) | Communication terminal and audio gain parameter regulation method | |
| CN101641965A (en) | Acoustic processing system and method for electronic device and mobile telephone terminal | |
| CN101123629A (en) | An environmental noise filtering method and device for mobile phone | |
| CN102223428A (en) | Noise reducing method and mobile terminal | |
| CN101140760A (en) | Sound signal collecting and processing system and method thereof | |
| CN103634439A (en) | Noise reduction processing system | |
| CN111885460A (en) | Transparent mode adjusting device and method of wireless earphone and wireless earphone | |
| US20090141906A1 (en) | Communication Headset Processing Multiple Audio Inputs | |
| US20130287203A1 (en) | Reduction of Loudspeaker Distortion for Improved Acoustic Echo Cancellation | |
| WO2023098401A1 (en) | Earphone having active noise reduction function and active noise reduction method | |
| JPH0490298A (en) | Hearing aid | |
| TW201933336A (en) | Electronic device and echo cancellation method applied to electronic device | |
| JP2016009935A (en) | Level adjustment circuit, digital sound processor, audio amplifier integrated circuit, electronic apparatus, and automatic level adjustment method of audio signal | |
| CN105474610A (en) | Acoustical signal processing method and device of communication device | |
| CN210694282U (en) | Digital conference system | |
| JP2000278786A (en) | Microphone system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150225 |