CN109243481A - A kind of method of speech processing and device - Google Patents
A kind of method of speech processing and device Download PDFInfo
- Publication number
- CN109243481A CN109243481A CN201811136340.1A CN201811136340A CN109243481A CN 109243481 A CN109243481 A CN 109243481A CN 201811136340 A CN201811136340 A CN 201811136340A CN 109243481 A CN109243481 A CN 109243481A
- Authority
- CN
- China
- Prior art keywords
- noise
- target
- noise model
- model
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012545 processing Methods 0.000 title claims abstract description 44
- 238000011946 reduction process Methods 0.000 claims abstract description 16
- 230000007613 environmental effect Effects 0.000 claims description 23
- 230000009467 reduction Effects 0.000 claims description 22
- 238000003860 storage Methods 0.000 claims description 22
- 230000033001 locomotion Effects 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 9
- 230000005236 sound signal Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 2
- 230000008030 elimination Effects 0.000 claims 1
- 238000003379 elimination reaction Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 12
- 238000004590 computer program Methods 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 235000015170 shellfish Nutrition 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009182 swimming Effects 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
- 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)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephone Function (AREA)
Abstract
The embodiment of the invention discloses a kind of method of speech processing and devices, are related to voice processing technology field, and main purpose is to improve voice de-noising efficiency；Main technical schemes are as follows: determine target motor pattern；The target motor pattern is included in preset at least one motor pattern；Each described motor pattern respectively corresponds an at least noise model；According to the ambient noise signal selection target noise model in current environment, the target noise model is included at least one corresponding noise model of the target motor pattern；Obtain voice signal；Noise reduction process is carried out to the voice signal using the target noise model.
Description
Technical field
The present embodiments relate to voice processing technology fields, more particularly to a kind of method of speech processing and device.
Background technique
The electronic equipments such as wearable device, wrist-watch are due to function such as music, video playing, motion state records
Can, therefore the user of electronic equipment would generally carry electronic equipment during exercise, so that electronic equipment increases in motion process
Entertainment.
Currently, user can carry out interactive voice with electronic equipment during the motion, in order to improve the effect of interactive voice,
By needing to carry out noise reduction process to voice, so that electronic equipment can correctly identify voice.The side of voice de-noising
Method is usually the diamylose gram noise reduction mode that uses, and passes through complicated sef-adapting filter noise reduction algorithm, Wiener filtering noise reduction algorithm
Scheduling algorithm carries out noise reduction.Since sef-adapting filter noise reduction algorithm, Wiener filtering noise reduction algorithm scheduling algorithm are complex, noise reduction mistake
It needs largely to be calculated in journey, therefore, the efficiency of noise reduction is lower.
Summary of the invention
In view of this, the embodiment of the present invention proposes a kind of method of speech processing and device, main purpose are to mention
High voice de-noising efficiency.
In a first aspect, the embodiment of the invention provides a kind of method of speech processing, which includes:
Determine target motor pattern；The target motor pattern is included in preset at least one motor pattern；It is each
The kind motor pattern respectively corresponds an at least noise model；
According to the ambient noise signal selection target noise model in current environment, the target noise model is included in institute
It states at least one corresponding noise model of target motor pattern；
Noise reduction process is carried out to voice signal using the target noise model.
It is optionally, described that noise reduction process is carried out to voice signal using the target noise model, comprising:
Inverse noise signal is generated using the target noise model；
It is superimposed the inverse noise signal and the voice signal, the voice signal after the noise that is eliminated.
Optionally, each described noise model respectively corresponds a sample energy grade；It is described according in current environment
Ambient noise signal selection target noise model, comprising:
Determine the corresponding environmental noise power grade of the ambient noise signal；
In at least one corresponding noise model of the target motor pattern, the target noise model is selected；Institute
It is identical to state environmental noise power grade sample energy grade corresponding with the target noise model.
Optionally it is determined that the corresponding environmental noise power grade of the ambient noise signal, comprising:
The corresponding environmental noise power grade of the ambient noise signal is determined using noise energy decision logic circuit；Institute
It states noise energy decision logic circuit and is set in advance in microphone output front end.
Optionally, each described noise model respectively corresponds an intensity of sound section；It is described according in current environment
Ambient noise signal selection target noise model, comprising:
Determine the corresponding target sound intensity of the ambient noise signal；
In at least one corresponding noise model of the target motor pattern, the target noise model is selected；Institute
Stating target sound intensity includes in the corresponding intensity of sound section of the target noise model.
Optionally, this method further include:
Judge whether current time executes new ambient noise collection period；
If executing, the new ambient noise signal in current environment is acquired；
According to the new ambient noise signal, judge whether to replace the target noise model；
If replacement, new target noise model is selected according to the new ambient noise signal.
Second aspect, the embodiment of the invention provides a kind of voice processing apparatus, which includes:
Determining module, for determining target motor pattern；The target motor pattern is included in preset at least one fortune
In dynamic model formula；Each described motor pattern respectively corresponds an at least noise model；
Selecting module, for according to the ambient noise signal selection target noise model in current environment, the target to be made an uproar
Acoustic model includes at least one corresponding noise model of the target motor pattern that the determining module determines；
Noise reduction module, the target noise model for being selected using the selecting module carry out noise reduction to voice signal
Processing.
Optionally, the noise reduction module, comprising:
Submodule is generated, for generating inverse noise signal using the target noise model；
It is superimposed submodule, for being superimposed the inverse noise signal and voice letter that the generation submodule generates
Number, the voice signal after the noise that is eliminated.
The third aspect, the embodiment of the invention provides a kind of storage medium, the storage medium includes the program of storage,
In, described program operation when control the storage medium where equipment execute it is any one of above-mentioned described in speech processes side
Method.
Fourth aspect includes processor, storage in the electronic equipment the embodiment of the invention provides a kind of electronic equipment
Device and bus；The processor, the memory complete mutual communication by the bus；The processor is for calling
Program instruction in the memory, with execute it is any one of above-mentioned described in method of speech processing.
The embodiment of the invention provides a kind of method of speech processing and devices, can be with after determining target motor pattern
Target noise model is selected according to the ambient noise signal in current environment, and target noise model is included in object run mould
In the corresponding noise model of formula.Noise reduction process is carried out to voice signal using target noise model.Since target noise model is
It is preset, when using target noise model to Speech processing, do not need to carry out complicated noise reduction operation, therefore,
Voice de-noising efficiency can be improved in scheme provided in an embodiment of the present invention.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 shows a kind of flow chart of method of speech processing provided by one embodiment of the present invention；
Fig. 2 shows a kind of flow charts for method of speech processing that another embodiment of the present invention provides；
Fig. 3 shows a kind of structural schematic diagram of voice processing apparatus provided by one embodiment of the present invention；
Fig. 4 shows a kind of structural schematic diagram of voice processing apparatus of another embodiment of the present invention offer；
Fig. 5 shows a kind of structural schematic diagram for voice processing apparatus that another embodiment of the invention provides；
Fig. 6 shows a kind of structural schematic diagram for voice processing apparatus that another embodiment of the invention provides；
Fig. 7 shows a kind of structural schematic diagram for voice processing apparatus that another embodiment of the invention provides；
Fig. 8 shows the structural schematic diagram of a kind of electronic equipment provided by one embodiment of the present invention.
Specific embodiment
It is described more fully the exemplary embodiment of the disclosure below with reference to accompanying drawings.Although showing this public affairs in attached drawing
The exemplary embodiment opened, it being understood, however, that may be realized in various forms the disclosure without the implementation that should be illustrated here
Example is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the model of the disclosure
It encloses and is fully disclosed to those skilled in the art.
As shown in Figure 1, the embodiment of the invention provides a kind of method of speech processing, which includes:
101, target motor pattern is determined；The target motor pattern is included in preset at least one motor pattern；
Each described motor pattern respectively corresponds an at least noise model；
102, according to the ambient noise signal selection target noise model in current environment, the target noise model includes
In at least one corresponding noise model of the target motor pattern；
103, noise reduction process is carried out to the voice signal using the target noise model.
Embodiment according to figure 1 can make an uproar after determining target motor pattern according to the environment in current environment
Acoustical signal selects target noise model, and target noise model is included in the corresponding noise model of object run mode.Benefit
Noise reduction process is carried out to voice signal with target noise model.Due to target noise model be it is preset, utilize target
When noise model is to Speech processing, do not need to carry out complicated noise reduction operation, therefore, scheme provided in an embodiment of the present invention
Voice de-noising efficiency can be improved.
In an embodiment of the invention, the method for speech processing can be applied according to business need wearable device,
In the electronic equipments such as mobile phone, portable computer and intelligent sound box.Wearable device can include but is not limited to wrist-watch, bracelet,
Earphone and glasses.
In an embodiment of the invention, the step 101 in flow chart shown in above-mentioned Fig. 1 determines the side of target motor pattern
At least there is the following two kinds in method:
The first, each motor pattern is respectively provided with the preset opening time, and different motor patterns is corresponding different
Opening time.Opening time can set according to specific business demand.It is executed respectively for each motor pattern: judgement
Whether current time reaches the motor pattern corresponding opening time；If so, the motor pattern is determined as target movement mould
Formula.
It second, will fortune corresponding with motor pattern mark when receiving the determine instruction for carrying motor pattern mark
Dynamic model formula is determined as target motor pattern.
In the present embodiment, the type of motor pattern can be determined according to business need.Optionally, according to existing all
Sports types sets motor pattern.For example, motor pattern can include but is not limited to indoor body-building mode, outdoor running
At least one of mode, swimming mode, ride mode, ball game mode.
In the present embodiment, each motor pattern respectively corresponds an at least noise model.Each noise model difference
Different noise signals is respectively included including a noise signal, and in the different corresponding noise models of motor pattern.Noise
Noise behavior when signal can be symbolized in motor pattern, in user's local environment.
In the present embodiment, each motor pattern respectively corresponds an at least noise model.The type of noise model can
To be determined according to business need.Optionally, at least there is the following two kinds type in noise model: the first, each noise model
Respectively correspond a sample energy grade.It should be noted that a kind of motor pattern corresponds at least one sample energy grade.The
Two kinds, each noise model respectively corresponds an intensity of sound section.It should be noted that a kind of motor pattern is corresponding each
A intensity of sound section is different.For example: the corresponding noise model 1 of motor pattern of riding and noise model 2, wherein make an uproar
Acoustic model 1 is corresponding intensity of sound section [0,30 decibel], and noise model 2 is corresponding intensity of sound section [31 decibels, 60 decibels].
In an embodiment of the invention, each described noise model involved in flow chart shown in above-mentioned Fig. 1 point
It Dui Ying not a sample energy grade；Step 102 in flow chart shown in above-mentioned Fig. 1 is believed according to the ambient noise in current environment
Number selection target noise model may include:
Determine the corresponding environmental noise power grade of the ambient noise signal；
In at least one corresponding noise model of the target motor pattern, the target noise model is selected；Institute
It is identical to state environmental noise power grade sample energy grade corresponding with the target noise model.
In the present embodiment, determine that the detailed process of the corresponding environmental noise power grade of ambient noise signal can wrap
It includes: the corresponding environmental noise power grade of ambient noise signal is determined using noise energy decision logic circuit；Noise energy is sentenced
Disconnected logic circuit is set in advance in microphone output front end.Each environmental noise power grade respectively corresponds an energy range
Between, and different environmental noise power grades corresponds to different energy sections.Environment is obtained in noise energy decision logic circuit
When noise signal, the corresponding energy of ambient noise signal can be identified, environment is then determined according to the energy identified
The corresponding environmental noise power grade of noise signal.Wherein it is determined that in the corresponding energy section of environmental noise power grade gone out
Including the energy identified.
In the present embodiment, after determining environmental noise power grade, in the corresponding noise mode of target motor pattern
Target noise model is selected in type.Wherein, ambient noise target noise model corresponding sample energy grade and determined
Energy grade is identical.
According to above-described embodiment, since each noise model respectively corresponds a sample energy grade, in determination
Out when the corresponding environmental noise power grade of ambient noise signal, can quickly it be selected according to environmental noise power grade
Target noise model, it is seen then that the selection speed of target noise model can be improved.
In an embodiment of the invention, each described noise model involved in flow chart shown in above-mentioned Fig. 1 point
It Dui Ying not an intensity of sound section；Step 102 in flow chart shown in above-mentioned Fig. 1 is believed according to the ambient noise in current environment
Number selection target noise model may include:
Determine the corresponding target sound intensity of the ambient noise signal；
In at least one corresponding noise model of the target motor pattern, the target noise model is selected；Institute
Stating target sound intensity includes in the corresponding intensity of sound section of the target noise model.
In the present embodiment, for example: the corresponding noise model 1 of target movement model " motor pattern of riding " and noise mode
Type 2, wherein noise model 1 is corresponding intensity of sound section [0,30 decibel], corresponding [31 points of the intensity of sound section of noise model 2
Shellfish, 60 decibels].Determine that the corresponding target sound intensity of ambient noise signal is 50 decibels, it is seen that 50 decibels to be included in sound strong
It spends in section [31 decibels, 60 decibels], selects noise model 2 for target noise model.
According to above-described embodiment, since each noise model respectively corresponds an intensity of sound section, in determination
Out when the corresponding intensity of sound of ambient noise signal, target noise model can be quickly selected according to intensity of sound, it can
See, the selection speed of target noise model can be improved.
In an embodiment of the invention, voice signal involved in the step 103 in flow chart shown in above-mentioned Fig. 1 is to use
The instruction voice that family is inputted to electronic equipment, so that electronic equipment can execute corresponding movement according to voice signal.
In the present embodiment, microphone can be opened in real time, and when there is voice signal input, microphone obtains language in real time
Sound signal.
In an embodiment of the invention, the step 103 in flow chart shown in above-mentioned Fig. 1 utilizes the target noise model
Noise reduction process is carried out to voice signal, may include:
Inverse noise signal is generated using the target noise model；
It is superimposed the inverse noise signal and the voice signal, the voice signal after the noise that is eliminated.
In the present embodiment, preset sample noise signal is respectively included in each noise model.
It may include: to be made an uproar according to target using the method that target noise model generates inverse noise signal in the present embodiment
The sample noise signal that acoustic model includes generates inverse noise signal equal with sample noise signal amplitude, opposite in phase.
In the present embodiment, be superimposed inverse noise signal and when voice signal, amplitude is identical, opposite in phase waveform can be
It disappears or reduces when superposition, to eliminate or become smaller the noise in voice signal.
According to above-described embodiment, inverse noise signal is generated using target noise model, is superimposed inverse noise signal and language
Sound signal, the voice signal after the noise that is eliminated.So as to eliminate the noise in voice signal to greatest extent, language is improved
The clarity of sound signal correctly identifies voice signal convenient for subsequent.
In an embodiment of the invention, which can also include the following steps:
Judge whether current time executes new ambient noise collection period；
If executing, the new ambient noise signal in current environment is acquired；
According to the new ambient noise signal, judge whether to replace the target noise model；
If replacement, new target noise model is selected according to the new ambient noise signal.
In the present embodiment, since the noise in user's local environment may generate variation over time,
Therefore in order to guarantee that the quality of noise reduction process can replace target noise model according to ambient noise signal.
In the present embodiment, ambient noise collection period can be set according to business need, for example, ambient noise is adopted
Collecting the period can be 10 minutes, 15 minutes, 30 minutes or 1 hour.Judging that current time can execute new ambient noise
When collection period, the new ambient noise signal in current environment is acquired.For example, being applied in wearable device in this method
When, new ambient noise signal can be acquired by the microphone in wearable device.The new ambient noise signal of acquisition
It can be moment a certain in new ambient noise collection period corresponding ambient noise signal, or, certain several moment corresponding ring
Border noise signal, wherein certain several moment can may be the discontinuous moment for the continuous moment.
In the present embodiment, according to new ambient noise signal, judge whether the method for replacement target noise model at least
Including following three kinds:
The first, new ambient noise signal quantity be one when, target motor pattern it is corresponding at least one
The corresponding noise model of new ambient noise signal is selected in noise model；Judge the noise model selected whether with currently
Target noise model it is identical, if not identical, judge to need replacing target noise model, and target noise model is replaced
For the noise model selected.If they are the same, then judge that current goal noise model does not need to replace.
It second, when the quantity of new ambient noise signal is multiple, finds out noise energy grade highest or sound is strong
Spend highest target environment noise signal.And it is selected at least one corresponding noise model of target motor pattern and target
The corresponding noise model of ambient noise signal；Judge whether the noise model selected is identical as current target noise model,
If not identical, judge to need replacing target noise model, and target noise model is changed to the noise model selected.
If they are the same, then judge that current goal noise model does not need to replace.
The third calculates being averaged for multiple ambient noise signals when the quantity of new ambient noise signal is multiple
Noise energy grade or average Voice intensity.And it selects and puts down at least one corresponding noise model of target motor pattern
Equal noise energy grade or the corresponding noise model of average Voice intensity；Judge the noise model selected whether with current mesh
It is identical to mark noise model, if not identical, judges to need replacing target noise model, and target noise model is changed to choosing
The noise model selected out.If they are the same, then judge that current goal noise model does not need to replace.
According to above-described embodiment, the noise in user's local environment may generate variation over time, because
This replaces target noise model according to ambient noise signal, allows target noise model with the variation of the noise in environment
And convert, the quality of noise reduction process can be improved.Based on the narration in the above embodiments, the embodiment of the above method can root
It needs to be freely combined according to business, is combined into new embodiment.
For being applied in wearable device " smartwatch " in this way below, method of speech processing is illustrated,
As shown in Fig. 2, the method for speech processing may include steps of:
201, target motor pattern is determined, wherein target motor pattern is included in preset at least one motor pattern；
Each motor pattern respectively corresponds at least a noise model, each noise model and respectively corresponds sample energy etc.
Grade.
In this step, there is ride motor pattern and road-work mode, wherein motion model of riding corresponds to noise mode
Type 1 and noise model 2, the corresponding sample energy grade 1 of noise model 1, the corresponding sample energy grade 2 of noise model 2.Road-work
Mode corresponds to noise model 3 and noise model 4, the corresponding sample energy grade 3 of noise model 3, the corresponding sample energy of noise model 4
Class 4.
In this step, when receiving the determine instruction for carrying motor pattern mark of riding, the motor pattern that will ride is true
It is set to target motor pattern.
202, determine that the corresponding environment of the ambient noise signal in current environment is made an uproar using noise energy decision logic circuit
Acoustic energy grade；Noise energy decision logic circuit is set in advance in microphone output front end.
In this step, ambient noise signal is obtained, ambient noise signal is determined using noise energy decision logic circuit
Corresponding environmental noise power grade is sample energy grade 1.
203, at least one corresponding noise model of target motor pattern, target noise model is selected；Environment is made an uproar
Acoustic energy grade sample energy grade corresponding with target noise model is identical.
In this step, the motion model that selects to ride corresponds to noise model 1 as target noise model.
204, voice signal is obtained.
In this step, voice signal is obtained, for example, the voice signal is the voice for allowing smartwatch to dial designated telephone
Signal.
205, inverse noise signal is generated using target noise model.
In this step, inverse noise signal is generated using target noise plant noise model 1.
206, inverse noise signal and voice signal, the voice signal after the noise that is eliminated are superimposed.
It in this step,, can since noise in speech signal is less after the voice signal after the noise that is eliminated
To carry out the identification of high quality to voice signal.
In the present embodiment, following steps can also be performed after determining target motor pattern: judges that current time is
It is no to execute new ambient noise collection period；If executing, the new ambient noise signal in current environment is acquired.According to new ring
Border noise signal judges whether to replace target noise model；If replacement, new target is selected according to new ambient noise signal
Noise model.If not executing, illustrates that current goal noise model can also meet noise reduction requirement, continue to use current goal and make an uproar
Acoustic model.
As shown in figure 3, the embodiment of the invention provides a kind of voice processing apparatus, which includes:
Determining module 301, for determining target motor pattern；The target motor pattern is included in preset at least one
In motor pattern；Each described motor pattern respectively corresponds an at least noise model；
Selecting module 302, for according to the ambient noise signal selection target noise model in current environment, the target
Noise model includes at least one corresponding noise model of the target motor pattern that the determining module determines；
Noise reduction module 303, for using the selecting module 302 selection the target noise model to voice signal into
Row noise reduction process.
Embodiment according to Fig.3, after determining module determines target motor pattern, selecting module can be according to working as
Ambient noise signal in preceding environment selects target noise model, and target noise model is included in object run mode and corresponds to
Noise model in.Noise reduction module carries out noise reduction process to voice signal using target noise model.Due to target noise model
Be it is preset, when using target noise model to Speech processing, do not need to carry out complicated noise reduction operation, because
This, voice de-noising efficiency can be improved in scheme provided in an embodiment of the present invention.
In an embodiment of the invention, as shown in figure 4, the noise reduction module 303 may include:
Submodule 3031 is generated, for generating inverse noise signal using the target noise model；
It is superimposed submodule 3032, the inverse noise signal speech signal generated for being superimposed the generation submodule,
Be eliminated the voice signal after noise.
In an embodiment of the invention, as shown in figure 5, each described noise model respectively corresponds a sample energy
Grade interval；The selecting module 302, comprising:
First determines submodule 3021, for determining the ambient noise signal corresponding environmental noise power grade area
Between；
First choice submodule 3022, for selecting at least one corresponding noise model of the target motor pattern
Select out the target noise model；Environmental noise power grade sample energy grade corresponding with the target noise model
It is identical.
In an embodiment of the invention, first submodule 3021 is determined, for using noise energy decision logic circuit
Determine the corresponding environmental noise power grade of the ambient noise signal；The noise energy decision logic circuit is set in advance in
Microphone exports front end.
In an embodiment of the invention, as shown in fig. 6, each described noise model respectively corresponds an intensity of sound
Section；The selecting module 302, comprising:
Second determines submodule 3023, for determining the corresponding target sound intensity of the ambient noise signal；
Second selection submodule 3024, for selecting at least one corresponding noise model of the target motor pattern
Select out the target noise model；The target sound intensity is included in the corresponding intensity of sound section of the target noise model
It is interior.
In an embodiment of the invention, as shown in fig. 7, the voice processing apparatus further include:
First judgment module 305, for judging whether current time executes new ambient noise collection period；If executing,
Triggering collection module 306；
Acquisition module 306, under the triggering of the first judgment module, the new environment acquired in current environment to be made an uproar
Acoustical signal；
Second judgment module 307, the new ambient noise signal for being acquired according to acquisition module 306, judgement are
The no replacement target noise model；If replacement, triggering selection module 302 selects newly according to the new ambient noise signal
Target noise model.
The contents such as information exchange, the implementation procedure between each module in above-mentioned apparatus, due to implementing with the method for the present invention
Example is based on same design, and for details, please refer to the description in the embodiment of the method for the present invention, and details are not described herein again.Based on above-mentioned
The embodiment of narration in embodiment, above-mentioned apparatus can need to be freely combined according to business, be combined into new embodiment.
The embodiment of the invention provides a kind of storage medium, the storage medium includes the program of storage, wherein described
Equipment program controls the storage medium when running where execute it is any one of above-mentioned described in environment temperature determine method.
The embodiment of the invention provides a kind of electronic equipment, as shown in figure 8, in the electronic equipment include processor 401,
Memory 402 and bus 403；The processor 401, the memory 402 complete mutual lead to by the bus 403
Letter；The processor 401 is used to call program instruction in the memory 402, with execute it is any one of above-mentioned described in
Environment temperature determines method.
It in embodiments of the present invention, can be according to the ambient noise in current environment after determining target motor pattern
Signal behavior goes out target noise model, and target noise model is included in the corresponding noise model of object run mode.It utilizes
Target noise model carries out noise reduction process to voice signal.Due to target noise model be it is preset, make an uproar using target
When acoustic model is to Speech processing, do not need to carry out complicated noise reduction operation, therefore, scheme provided in an embodiment of the present invention can
To improve voice de-noising efficiency.
In embodiments of the present invention, since each noise model respectively corresponds a sample energy grade, true
When making the corresponding environmental noise power grade of ambient noise signal, can quickly it be selected according to environmental noise power grade
Target noise model out, it is seen then that the selection speed of target noise model can be improved.
In embodiments of the present invention, since each noise model respectively corresponds an intensity of sound section, true
When making the corresponding intensity of sound of ambient noise signal, target noise model can be quickly selected according to intensity of sound,
As it can be seen that the selection speed of target noise model can be improved.
In embodiments of the present invention, using target noise model generate inverse noise signal, superposition inverse noise signal and
Voice signal, the voice signal after the noise that is eliminated.So as to eliminate the noise in voice signal to greatest extent, improve
The clarity of voice signal correctly identifies voice signal convenient for subsequent.
In embodiments of the present invention, the noise in user's local environment may generate variation over time,
Therefore target noise model is replaced according to ambient noise signal, allow target noise model with the change of the noise in environment
Change and convert, the quality of noise reduction process can be improved.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/
Or the forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable Jie
The example of matter.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including element
There is also other identical elements in process, method, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can provide as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application
Form.It is deposited moreover, the application can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The above is only embodiments herein, are not intended to limit this application.To those skilled in the art,
Various changes and changes are possible in this application.It is all within the spirit and principles of the present application made by any modification, equivalent replacement,
Improve etc., it should be included within the scope of the claims of this application.
Claims (10)
1. a kind of method of speech processing characterized by comprising
Determine target motor pattern；The target motor pattern is included in preset at least one motor pattern；Each institute
It states motor pattern and respectively corresponds an at least noise model；
According to the ambient noise signal selection target noise model in current environment, the target noise model is included in the mesh
It marks at least one corresponding noise model of motor pattern；
Noise reduction process is carried out to voice signal using the target noise model.
2. method of speech processing according to claim 1, which is characterized in that described to utilize the target noise model to language
Sound signal carries out noise reduction process, comprising:
Inverse noise signal is generated using the target noise model；
It is superimposed the inverse noise signal and the voice signal, the voice signal after the noise that is eliminated.
3. method of speech processing according to claim 1 or 2, which is characterized in that each described noise model is right respectively
Answer a sample energy grade；The ambient noise signal selection target noise model according in current environment, comprising:
Determine the corresponding environmental noise power grade of the ambient noise signal；
In at least one corresponding noise model of the target motor pattern, the target noise model is selected；The ring
Noise energy grade sample energy grade corresponding with the target noise model in border is identical.
4. method of speech processing according to claim 3, which is characterized in that determine the corresponding ring of the ambient noise signal
Border noise energy grade, comprising:
The corresponding environmental noise power grade of the ambient noise signal is determined using noise energy decision logic circuit；It is described to make an uproar
Acoustic energy decision logic circuit is set in advance in microphone output front end.
5. method of speech processing according to claim 1 or 2, which is characterized in that each described noise model is right respectively
Answer an intensity of sound section；The ambient noise signal selection target noise model according in current environment, comprising:
Determine the corresponding target sound intensity of the ambient noise signal；
In at least one corresponding noise model of the target motor pattern, the target noise model is selected；The mesh
Marking intensity of sound includes in the corresponding intensity of sound section of the target noise model.
6. method of speech processing according to claim 1 or 2, which is characterized in that this method further include:
Judge whether current time executes new ambient noise collection period；
If executing, the new ambient noise signal in current environment is acquired；
According to the new ambient noise signal, judge whether to replace the target noise model；
If replacement, new target noise model is selected according to the new ambient noise signal.
7. a kind of voice processing apparatus characterized by comprising
Determining module, for determining target motor pattern；The target motor pattern is included in preset at least one movement mould
In formula；Each described motor pattern respectively corresponds an at least noise model；
Selecting module, for according to the ambient noise signal selection target noise model in current environment, the target noise mould
Type includes at least one corresponding noise model of the target motor pattern that the determining module determines；
Noise reduction module, the target noise model for being selected using the selecting module carry out at noise reduction voice signal
Reason.
8. voice processing apparatus according to claim 7, which is characterized in that the noise reduction module, comprising:
Submodule is generated, for generating inverse noise signal using the target noise model；
It is superimposed submodule, the inverse noise signal and the voice signal generated for being superimposed the generation submodule obtains
Voice signal to after elimination noise.
9. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein run in described program
When control the storage medium where equipment perform claim require 1 to the speech processes side described in any one of claim 6
Method.
10. a kind of electronic equipment, which is characterized in that include processor, memory and bus in the electronic equipment；The processing
Device, the memory complete mutual communication by the bus；The processor is used to call the journey in the memory
Sequence instruction, with method of speech processing described in any one of perform claim requirement 1 to claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811136340.1A CN109243481A (en) | 2018-09-28 | 2018-09-28 | A kind of method of speech processing and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811136340.1A CN109243481A (en) | 2018-09-28 | 2018-09-28 | A kind of method of speech processing and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109243481A true CN109243481A (en) | 2019-01-18 |
Family
ID=65057810
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811136340.1A Pending CN109243481A (en) | 2018-09-28 | 2018-09-28 | A kind of method of speech processing and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109243481A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111833895A (en) * | 2019-04-23 | 2020-10-27 | 北京京东尚科信息技术有限公司 | Audio signal processing method, apparatus, computer device and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120201393A1 (en) * | 2011-02-04 | 2012-08-09 | Cardo Systems, Inc. | System and method for adjusting audio input and output settings |
| CN104064185A (en) * | 2013-03-18 | 2014-09-24 | 联想(北京)有限公司 | Information processing method and system and electronic device |
| CN104378774A (en) * | 2013-08-15 | 2015-02-25 | 中兴通讯股份有限公司 | Voice quality processing method and device |
| CN105657150A (en) * | 2015-09-29 | 2016-06-08 | 宇龙计算机通信科技(深圳)有限公司 | Noise elimination method and device and electronic device |
| CN106297779A (en) * | 2016-07-28 | 2017-01-04 | 块互动（北京）科技有限公司 | A kind of background noise removing method based on positional information and device |
-
2018
- 2018-09-28 CN CN201811136340.1A patent/CN109243481A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120201393A1 (en) * | 2011-02-04 | 2012-08-09 | Cardo Systems, Inc. | System and method for adjusting audio input and output settings |
| CN104064185A (en) * | 2013-03-18 | 2014-09-24 | 联想(北京)有限公司 | Information processing method and system and electronic device |
| CN104378774A (en) * | 2013-08-15 | 2015-02-25 | 中兴通讯股份有限公司 | Voice quality processing method and device |
| CN105657150A (en) * | 2015-09-29 | 2016-06-08 | 宇龙计算机通信科技(深圳)有限公司 | Noise elimination method and device and electronic device |
| CN106297779A (en) * | 2016-07-28 | 2017-01-04 | 块互动（北京）科技有限公司 | A kind of background noise removing method based on positional information and device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111833895A (en) * | 2019-04-23 | 2020-10-27 | 北京京东尚科信息技术有限公司 | Audio signal processing method, apparatus, computer device and medium |
| CN111833895B (en) * | 2019-04-23 | 2023-12-05 | 北京京东尚科信息技术有限公司 | Audio signal processing method, device, computer equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106445460B (en) | Control method and device | |
| CN109120983A (en) | A kind of audio-frequency processing method and device | |
| CN108269569A (en) | Audio recognition method and equipment | |
| CN106302678A (en) | A kind of music recommends method and device | |
| CN106328126A (en) | Far-field speech recognition processing method and device | |
| CN108920510A (en) | Automatic chatting method, device and electronic equipment | |
| CN111325444B (en) | Risk prevention and control decision method, device, system and equipment | |
| CN112614504A (en) | Single sound channel voice noise reduction method, system, equipment and readable storage medium | |
| CN109389976A (en) | Intelligent household electrical appliance control method and device, intelligent household electrical appliance and storage medium | |
| CN110392302A (en) | Video is dubbed in background music method, apparatus, equipment and storage medium | |
| CN113643714B (en) | Audio processing method, device, storage medium and computer program | |
| CN108810746A (en) | A kind of sound quality optimization method, feedback noise reduction system, earphone and storage medium | |
| CN104317392B (en) | A kind of information control method and electronic equipment | |
| CN109658935A (en) | The generation method and system of multichannel noisy speech | |
| CN110456309A (en) | Sound localization method, device and computer readable storage medium | |
| CN108831459A (en) | Audio recognition method and device | |
| CN109243481A (en) | A kind of method of speech processing and device | |
| CN106131472A (en) | A kind of kinescope method and mobile terminal | |
| CN105516860A (en) | Virtual basetone generating method, virtual basetone generating device and terminal | |
| CN107820677A (en) | Determine the method and its device, terminal of filter coefficient | |
| CN112992170B (en) | Model training method and device, storage medium and electronic device | |
| CN117765142A (en) | Digital person generating method, device and digital person generating system | |
| CN111883173B (en) | Audio packet loss repairing method, equipment and system based on neural network | |
| CN117369765A (en) | Equalizer self-adaptive adjusting method and related equipment | |
| CN107247574A (en) | A kind of audio output control method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |
|
| RJ01 | Rejection of invention patent application after publication |