CN107369441A - Noise-eliminating method, device and the terminal of voice signal - Google Patents
Noise-eliminating method, device and the terminal of voice signal Download PDFInfo
- Publication number
- CN107369441A CN107369441A CN201710810085.3A CN201710810085A CN107369441A CN 107369441 A CN107369441 A CN 107369441A CN 201710810085 A CN201710810085 A CN 201710810085A CN 107369441 A CN107369441 A CN 107369441A
- Authority
- CN
- China
- Prior art keywords
- voice signal
- mic
- terminal
- noising
- voice
- 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 87
- 238000012545 processing Methods 0.000 claims abstract description 48
- 230000009467 reduction Effects 0.000 claims abstract description 46
- 230000009977 dual effect Effects 0.000 claims abstract description 35
- 230000000052 comparative effect Effects 0.000 claims abstract description 16
- 230000001755 vocal effect Effects 0.000 claims description 29
- 230000008569 process Effects 0.000 claims description 15
- 238000013075 data extraction Methods 0.000 claims description 10
- 230000005236 sound signal Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 description 17
- 230000000694 effects Effects 0.000 description 7
- 238000000605 extraction Methods 0.000 description 7
- 238000012544 monitoring process Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000011410 subtraction method Methods 0.000 description 2
- 102000008482 12E7 Antigen Human genes 0.000 description 1
- 108010020567 12E7 Antigen Proteins 0.000 description 1
- 101000893549 Homo sapiens Growth/differentiation factor 15 Proteins 0.000 description 1
- 101000692878 Homo sapiens Regulator of MON1-CCZ1 complex Proteins 0.000 description 1
- 102100026436 Regulator of MON1-CCZ1 complex Human genes 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000001960 triggered effect 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
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- 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
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
-
- 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
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- 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
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L2021/02161—Number of inputs available containing the signal or the noise to be suppressed
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Telephone Set Structure (AREA)
Abstract
The embodiment of the present invention provides a kind of noise-eliminating method of voice signal, device and terminal, is related to voice de-noising technical field, this method includes：Voice signal is received by the MIC of terminal, wherein, voice signal includes：First MIC the first voice signal and the 2nd MIC the second voice signal；When needing to carry out the selection of de-noising pattern, compare the volume of the first voice signal and the second voice signal；De-noising pattern is chosen according to comparative result, de-noising pattern includes：Dual microphone noise reduction mode and steady-state noise tupe；Denoising Processing is carried out to the voice signal received according to the de-noising pattern of selection.Noise-eliminating method, device and the terminal of voice signal provided in an embodiment of the present invention, de-noising pattern is chosen on the basis of the volume for the voice signal that two MIC are received is compared, the de-noising pattern of selection is set to be more suitable for current MIC actual performance, therefore the overall noise reduction of terminal is improved, improves user experience.
Description
Technical field
The present invention relates to voice de-noising technical field, in particular to a kind of noise-eliminating method of voice signal, device and
Terminal.
Background technology
At present, mainly have to the noise reduction mode of voice signal during call or voice recording：It is whole for dual microphone
Dual microphone (Microphone, the abridge MIC) noise reduction at end and the single microphone noise reduction for single microphone terminal.
Single microphone noise reduction mode is poor to the adaptability of environmental change, it is necessary to gather related letter when conversing and starting
Breath, and when user's carried terminal carries out position movement, ambient noise also changes therewith, causes original noise to eliminate
Be not suitable for current environment, de-noising effect is poor；
The mode of dual microphone noise reduction, main microphon pickup voice, auxiliary microphone pick up ambient noise, pass through noise reduction algorithm
Reach the effect for eliminating noise and improving call, it is but higher to the quality requirement of microphone, such as exist in microphone airtight
During the problems such as property, situations such as small In Call or audio discontinuity can be caused.
For above-mentioned dual microphone noise reduction mode, the problem of causing de-noising effect poor because of microphone problem, at present not yet
It is proposed effective solution.
The content of the invention
In view of this, the purpose of the embodiment of the present invention is to provide a kind of noise-eliminating method of voice signal, device and terminal,
To lift the de-noising effect of dual microphone terminal.
To achieve these goals, the technical scheme that the embodiment of the present invention uses is as follows：
In a first aspect, the embodiments of the invention provide a kind of noise-eliminating method of voice signal, applied at least two MIC's
Terminal, this method include：Voice signal is received by the MIC of terminal, wherein, voice signal includes：First MIC the first voice
Second voice signal of signal and the 2nd MIC；When needing to carry out the selection of de-noising pattern, compare the first voice signal and the second language
The volume of sound signal；De-noising pattern is chosen according to comparative result, de-noising pattern includes：Dual microphone noise reduction mode and stable state
Noise processed pattern；Denoising Processing is carried out to the voice signal received according to the de-noising pattern of selection.
In preferred embodiments of the present invention, the above method also includes：Judge the first voice signal and the second voice signal
In whether include the key message that prestores；If it is, determine to need to carry out de-noising pattern selection.
In preferred embodiments of the present invention, above by terminal MIC receive voice signal the step of include：Work as terminal
When starting voice signal receive capabilities, re-reading prompt message is shown in the display interface of terminal, re-reading prompt message includes re-reading
The key message to prestore；Voice signal corresponding to the key message to prestore is received by the MIC of terminal.
In preferred embodiments of the present invention, the above method also includes：When default judgement is identified as the first setting identification
When, it is determined that needing to carry out de-noising pattern selection；When the de-noising pattern of selection is steady-state noise tupe, sets and judge mark
For default second setting identification, and prompt the user with de-noising pattern.
In preferred embodiments of the present invention, the volume of the above-mentioned voice signal of comparison first and the second voice signal
Step includes：The first energy of the first voice signal and the second energy of the second voice signal are obtained, using the first energy as
The volume of one voice signal, the volume using the second energy as the second voice signal；Or obtain the first of the first voice signal
Second intensity of intensity and the second voice signal, the volume using the first intensity as the first voice signal, using the second intensity as
The volume of second voice signal.
In preferred embodiments of the present invention, above-mentioned the step of choosing de-noising pattern according to comparative result, includes：When first
MIC is main MIC, and when the volume of the first voice signal subtracts the difference of the volume of the second voice signal and is more than given threshold, choosing
Take dual microphone noise reduction mode；Otherwise, steady-state noise tupe is chosen.
In preferred embodiments of the present invention, the above method also includes：When receiving the first voice signal and the second voice
During signal, the first voice signal and the second voice signal are replicated；The first voice signal and the second voice signal replicated is superimposed, it is raw
Into voice superposed signal；Voice vocal print feature is extracted from speech superposition signal；De-noising pattern according to selection is to receiving
After voice signal carries out the step of denoising Processing, the above method also includes：According to voice vocal print feature, after denoising Processing
Voice signal carries out voice data extraction, obtains final speech data.
In preferred embodiments of the present invention, the above method also includes：If voice signal is during telephone relation
Voice signal, the equipment that above-mentioned final speech data is sent to call opposite end；If voice signal is the language during video
Sound signal, the equipment that the image in above-mentioned final speech data and video is sent to video opposite end；If voice signal is record
Speech data during sound, above-mentioned final speech data is saved as into recorded audio.
Second aspect, the embodiments of the invention provide a kind of noise cancellation apparatus of voice signal, applied at least two MIC's
Terminal, the device include：Voice signal receiving module, for receiving voice signal by the MIC of terminal, wherein, voice signal
Including：First MIC the first voice signal and the 2nd MIC the second voice signal；Volume comparison module, need to carry out for working as
When de-noising pattern is chosen, compare the volume of the first voice signal and the second voice signal；De-noising pattern chooses module, is used for
De-noising pattern is chosen according to comparative result, de-noising pattern includes：Dual microphone noise reduction mode and steady-state noise tupe；De-noising
Processing module, denoising Processing is carried out to the voice signal received for the de-noising pattern according to selection.
In preferred embodiments of the present invention, said apparatus also includes：Key message judge module, for judging the first language
Whether the key message to prestore is included in sound signal and the second voice signal；If it is, determine to need to carry out de-noising pattern selection.
In preferred embodiments of the present invention, above-mentioned voice signal receiving module is additionally operable to：When terminal starts voice signal
During receive capabilities, re-reading prompt message is shown in the display interface of terminal, re-reading prompt message includes the re-reading crucial letter to prestore
Breath；Voice signal corresponding to the key message to prestore is received by the MIC of terminal.
In preferred embodiments of the present invention, said apparatus also includes：De-noising pattern identification module, for that default ought sentence
It is disconnected when being identified as the first setting identification, it is determined that needing to carry out de-noising pattern selection；When the de-noising pattern of selection is at steady-state noise
During reason pattern, judgement is set to be identified as default second setting identification, and prompt the user with de-noising pattern.
In preferred embodiments of the present invention, above-mentioned volume comparison module is additionally operable to：Obtain the first of the first voice signal
Second energy of energy and the second voice signal, the volume using the first energy as the first voice signal, using the second energy as
The volume of second voice signal；Or the first intensity of the first voice signal and the second intensity of the second voice signal are obtained, will
Volume of first intensity as the first voice signal, the volume using the second intensity as the second voice signal.
In preferred embodiments of the present invention, above-mentioned de-noising pattern is chosen module and is additionally operable to：When the first MIC is main MIC, and
When the difference that the volume of first voice signal subtracts the volume of the second voice signal is more than given threshold, dual microphone noise reduction is chosen
Pattern；Otherwise, steady-state noise tupe is chosen.
In preferred embodiments of the present invention, said apparatus also includes：Vocal print feature extraction module, for when receiving the
When one voice signal and the second voice signal, the first voice signal and the second voice signal are replicated；It is superimposed the first voice replicated
Signal and the second voice signal, generate speech superposition signal；Voice vocal print feature is extracted from speech superposition signal；Voice data
Extraction module, for according to voice vocal print feature, carrying out voice data extraction to the voice signal after denoising Processing, obtaining final
Speech data.
In preferred embodiments of the present invention, said apparatus also includes：Language data process module, if believed for voice
Number be telephone relation during voice signal, by final voice signal be sent to call opposite end equipment；If voice signal
For the voice signal during video, the equipment that the image in final speech data and video is sent to video opposite end；If
Voice signal is the voice signal in Recording Process, and final speech data is saved as into recorded audio.
The third aspect, the embodiments of the invention provide a kind of terminal, including memory and processor, memory is used to deposit
The program of any one of computing device first aspect method is supported in storage, and processor is configurable for performing what is stored in memory
Program.
Fourth aspect, the embodiments of the invention provide a kind of computer-readable storage medium, for saving as above-mentioned second aspect
Computer software instructions used in any one device.
Compared with prior art, the noise-eliminating method of voice signal provided in an embodiment of the present invention, device and terminal, pass through end
First MIC at end receives the first voice signal and the 2nd MIC and receives the second voice signal, when needing to carry out de-noising pattern selection
When, compare the volume of the first voice signal and the second voice signal, and de-noising pattern, this side are chosen according to comparative result
Formula is selected because terminal is configured with two kinds of de-noising patterns on the basis of the volume for the voice signal that two MIC are received is compared
Cancel pattern of making an uproar, the de-noising pattern of selection is more suitable for current MIC actual performance, therefore improve the overall noise reduction of terminal
Effect, improve user experience.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by embodiment it is required use it is attached
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, therefore be not construed as pair
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, can also be according to this
A little accompanying drawings obtain other related accompanying drawings.
Fig. 1 is a kind of structured flowchart of terminal provided in an embodiment of the present invention；
Fig. 2 is a kind of flow chart of the noise-eliminating method of voice signal provided in an embodiment of the present invention；
Fig. 3 is a kind of block schematic illustration of terminal provided in an embodiment of the present invention；
Fig. 4 is a kind of particular flow sheet of the noise-eliminating method of voice signal provided in an embodiment of the present invention；
Fig. 5 is the schematic flow sheet of steady-state noise tupe provided in an embodiment of the present invention；
Fig. 6 is the principle schematic of dual microphone noise reduction mode provided in an embodiment of the present invention；
Fig. 7 is the structured flowchart of the noise cancellation apparatus of the first voice signal provided in an embodiment of the present invention；
Fig. 8 is the structured flowchart of the noise cancellation apparatus of second of voice signal provided in an embodiment of the present invention；
Fig. 9 is the structured flowchart of the noise cancellation apparatus of the third voice signal provided in an embodiment of the present invention；
Figure 10 is the structured flowchart of the noise cancellation apparatus of the 4th kind of voice signal provided in an embodiment of the present invention；
Figure 11 is the structured flowchart of the noise cancellation apparatus of the 5th kind of voice signal provided in an embodiment of the present invention.
Embodiment
Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.Generally exist
The component of the embodiment of the present invention described and illustrated in accompanying drawing can be configured to arrange and design with a variety of herein.Cause
This, the detailed description of the embodiments of the invention to providing in the accompanying drawings is not intended to limit claimed invention below
Scope, but it is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, those skilled in the art are not doing
The every other embodiment obtained on the premise of going out creative work, belongs to the scope of protection of the invention.
The following each embodiments of the present invention can be applied to the terminal for including at least two MIC.The terminal can be mobile whole
End, sound pick-up outfit, video call device, tablet personal computer, PC, vehicle-mounted computer, Intelligent worn device etc. arbitrarily need into
The terminal device of row speech denoising, it can apply to converse, record, video calling etc. is related to the application of voice.Fig. 1 shows one
The structured flowchart of kind terminal provided in an embodiment of the present invention, the terminal 100 include：Radio frequency (Radio Frequency, RF) circuit
110th, memory 120, input block 130, display unit 140, sensor 150, voicefrequency circuit 160, Wireless Fidelity (wireless
Fidelity, WiFi) part such as module 170, processor 180 and power supply 190.Wherein, voicefrequency circuit 160 is connected with least
Two MIC.It will be understood by those skilled in the art that the restriction of the structure of terminal 100 shown in Fig. 1 not structure paired terminal 100,
It can include than illustrating more or less parts, either combine some parts or different parts arrangement.
The embodiments of the invention provide a kind of noise-eliminating method of voice signal, device and terminal；The technology can apply to
Include at least two MIC terminal, such as the terminal shown in above-mentioned Fig. 1；The technology can use related software or hardware real
It is existing, it is described below by embodiment.
Embodiment one
A kind of flow chart of the noise-eliminating method of voice signal shown in Figure 2, this method are applied to include at least two
MIC terminal, this method comprise the following steps：
Step S201, voice signal is received by the MIC of terminal.
Wherein, voice signal includes：First MIC the first voice signal and the 2nd MIC the second voice signal.With terminal
Exemplified by being mobile phone, the first MIC and the 2nd MIC type selecting and require to carry out according to each platform vendor in the position that terminal is placed, example
Being approached as two MIC are designed from model of the same race or audio cavity structure, the first MIC is also referred to as the MIC that decides and is typically placed in mobile phone bottom,
2nd MIC is also referred to as de-noising MIC or auxiliary MIC, can be placed on the top at mobile phone top or the mobile phone back side.
Conversed or recorded etc. be related to the operation of voice when, voice letter is received by the first MIC and the 2nd MIC respectively
Number, the voice signal includes voice and ambient noise simultaneously.
Step S202, when needing to carry out the selection of de-noising pattern, compare the sound of the first voice signal and the second voice signal
Measure size.
De-noising pattern includes dual microphone noise reduction mode and steady-state noise tupe, the steady-state noise in the present embodiment
Tupe reduces noise based on single microphone combination vocal print system, and it is former that the dual microphone noise reduction mode is based on energy subtraction method
Superposition is managed to reduce noise.It can be chosen according to Mike's wind regime of present terminal, especially by the sound of voice signal
Size is measured to determine.
Herein it should be noted that for the reasonability of lifting scheme, avoid, to all voice signals received, entering
The selection of row de-noising pattern, when receiving voice signal, the present embodiment also needs to first determine the need for carrying out de-noising pattern
Choose., can be poor by calculating the volume of the first voice signal and the second voice signal if necessary to carry out de-noising pattern selection
Value determines the size of both volumes.
Step S203, de-noising pattern is chosen according to comparative result.
Specifically, de-noising mould is determined according to the result of above-mentioned first voice signal and the volume difference of the second voice signal
Formula, the difference represent each Mike's wind regime of present terminal.For example, it is main MIC to work as the first MIC, and the first voice is believed
Number the difference of the volume volume that subtracts the second voice signal when being more than given threshold, represent now each microphone and be in just
Normal state, choose dual microphone noise reduction mode；Otherwise, steady-state noise tupe is chosen.If it is not in just in microphone
Dual microphone noise reduction mode is used during normal state, terminal speech function can be caused to be unable to normal use, for example, call sound it is small or
The problems such as person's call is interrupted.
Step S204, denoising Processing is carried out to the voice signal received according to the de-noising pattern of selection.Terminal is according to choosing
Denoising Processing flow corresponding to the de-noising pattern execution taken.
The above method that the present embodiment provides, the first voice signal is received by the first MIC of terminal and the 2nd MIC is received
Second voice signal, when needing to carry out de-noising pattern to choose, compare giving great volume for the first voice signal and the second voice signal
It is small, and de-noising pattern is chosen according to comparative result, this mode is comparing two because terminal is configured with two kinds of de-noising patterns
De-noising pattern is chosen on the basis of the volume for the voice signal that MIC is received, the de-noising pattern of selection is more suitable for currently
MIC actual performance, therefore the overall noise reduction of terminal is improved, improve user experience.
It can be arranged as required to for the specific implementation for determining the need for carrying out de-noising pattern selection, such as：
Whether judge in the first voice signal and the second voice signal comprising the key message to prestore；If it is, determine that needs are disappeared
Pattern of making an uproar is chosen.
Comprising the key message to prestore in first voice signal and the second voice signal, one below can be included：(1)
Comprising the key message to prestore in one voice signal, and also comprising the key message that this prestores in the second voice signal；(2) first
Comprising the key message to prestore in voice signal, or, the key message that this prestores is included in the second voice signal.
The key message to prestore can be by system configuration, can also be by user configuration, voice work(of the user in using terminal
During energy, voice signal corresponding to the key message that this prestores can be provided, and then trigger the selection flow of de-noising pattern.Such as：
In the phonetic function of user's using terminal, terminal can show that brief conventional term is re-reading or using voice for user
Flexibly said during function.Such as the conventional term made a phone call includes：" feed, you are good, and who are you", " hello, and which you are", " you
It is good, it may I ask that who are you" etc..Above-mentioned conventional term is pre-set, such as is stored in advance in the RAM of terminal.Terminal can identify language
The key wordses information that system is set whether is included in sound signal, is triggered by key wordses information and carries out de-noising pattern selection.
Or above-mentioned the step of determining the need for carrying out the selection of de-noising pattern, it can also judge that mark is real by one
It is existing, it is above-mentioned to determine the need for include the step of carrying out the selection of de-noising pattern based on this：When default judgement is identified as
During the first setting identification, it is determined that needing to carry out de-noising pattern selection；When the de-noising pattern of selection is steady-state noise tupe,
Set judgement to be identified as default second setting identification, and prompt the user with the de-noising pattern (i.e. steady-state noise tupe).
The default identification that first setting identification is set when can be and dispatch from the factory, such as be arranged to 0 when dispatching from the factory, now acquiescence is eventually
The microphone at end is all in good condition, that is, meets the hardware condition for performing dual microphone noise reduction mode.
When the de-noising pattern that the comparison step by the above method determines to choose is steady-state noise tupe, show this
When terminal microphone problem be present, be unsatisfactory for perform dual microphone noise reduction mode hardware condition, de-noising afterwards is not
The selection (the steady-state noise tupe based on single microphone can only be used) for carrying out tupe again is needed to need to will determine that mark
Knowledge is revised as default second setting identification.Second setting identification is the mark different from first setting identification, such as is set
It is set to 1.The selection result due to it the talking mode of user's handheld terminal be present caused by is steady-state noise tupe, also
The steady-state noise tupe can be prompted the user with, user manually can carry out judging the modification of mark.When
When carrying out microphone renewal, factory reset.
For the judgment mode of above-mentioned key message, above-mentioned steps S201 can also include：When terminal starts voice signal
During receive capabilities, re-reading prompt message is shown in the display interface of terminal, re-reading prompt message includes the re-reading crucial letter to prestore
Breath；Voice signal corresponding to the key message to prestore is received by the MIC of terminal.
As a kind of embodiment, the volume of the first voice signal and the second voice signal in above-mentioned steps S202 can be with
Determined by one of following two modes：
(1) obtain the first voice signal the first energy and the second voice signal the second energy, using the first energy as
The volume of first voice signal, the volume using the second energy as the second voice signal；
(2) obtain the first voice signal the first intensity and the second voice signal the second intensity, using the first intensity as
The volume of first voice signal, the volume using the second intensity as the second voice signal.
Whether the above method that the present embodiment provides, can be by including the key message or end to prestore in voice signal
The judgement mark of end configuration determines whether that triggering carries out de-noising pattern selection, improves the flexibility of triggering, according to the above method,
Can be complicated and change to adapt to ambient noise using dual microphone noise reduction mode when the microphone of terminal is in normal condition
In the case of noise signal elimination, not being moved to noisy environment from quiet environment by terminal is influenceed；At the microphone of terminal
When abnormal condition, steady-state noise tupe can also be used to carry out noise elimination, ensure that voice to a certain extent
The audio of signal and voice quality is optimized, improve user experience.
Embodiment two
Fig. 3 shows the block schematic illustration of terminal provided in an embodiment of the present invention.The terminal of the present embodiment includes microphone
MIC 1 and MIC 2, key message monitoring module 3, vocal print processing module 4, comparator 5, noise processed module 6 and speech data
Acquisition module 7.
Wherein, MIC 1 and MIC 2 are connected with key message monitoring module 3 and noise processed module 6 respectively, key message
Monitoring module 3 is also connected with vocal print processing module 4 and comparator 5, and noise processed module 6 also obtains with comparator 5 and speech data
Modulus block 7 is connected, and the other end of vocal print processing module 4 is connected with speech data acquisition module 7；The function of modules is as follows：
(1) MIC 1 and MIC 2 receive voice signal respectively；
(2) whether key message monitoring module 3 is used to identify pre- comprising system in the voice signal in MIC 1 and MIC 2
If key message (such as the information such as key wordses, key sentence), when recognizing key message, to MIC's 1 and MIC 2
Voice signal is extracted, and all the way the voice signal with key message is not done into any speech processes exports to comparator, separately
Exported after MIC 1 and MIC 2 voice signal are superimposed all the way outside to vocal print processing module.
(3) MIC 1 and 2 respective voice signals of MIC that comparator 5 is used to be exported according to key message monitoring module enter
Row volume ratio is compared with the present embodiment by taking energy as an example, the present embodiment compares MIC 1 voice signal-MIC 2 voice signal
≥6dB。
(4) noise processed module 6 is used to carry out de-noising pattern selection according to the comparative result of comparator, as shown in Figure 3
The concrete structure of noise processed module 6, the noise processed module include DSP (Digital Signal Process, data signal
Processing) chip 61, dual microphone noise reduction processing unit 62 and stable state noise reduction processing unit 63, concrete function is as follows：
As MIC 1 voice signal-MIC 2 voice signal >=6dB；The now dsp chip inside noise processed module 6
61 judge that mobile phone state is normal, can start dual microphone noise reduction processing unit 62, while stable state noise reduction processing unit is closed.
Dual microphone noise reduction processing unit mainly runs algorithm corresponding to dual microphone noise reduction mode.
When the key message signals of 1 key message signal-MIC of MIC 2<6dB, now noise processed inside modules DSP cores
Piece 61 judges that mobile phone state is abnormal, it is understood that there may be the cavity air-tightness leakage of primary/secondary microphone or monomer microphone are present
Bad phenomenon, it is impossible to start dual microphone noise reduction processing unit 62, otherwise can have other side and hear that sound is small or audio discontinuity etc.
Situation, influence normal talking；Now can triggering stable state noise reduction processing unit 63 works, and stable state noise reduction processing unit 63 is main
Processing Algorithm corresponding to steady-state noise tupe is run, can be handled just for the sound picked up of MIC 1.
Voice messaging transmission after above-mentioned noise processed module 6 will have been handled is given to speech data acquisition module 7.
(5) vocal print processing module 4 is used for the language for being superimposed with MIC 1 and MIC 2 for receiving the output of key message monitoring module 3
Sound signal, voice vocal print feature extraction is carried out to the voice signal, and vocal print feature extraction is typically Frequency spectrum ratio to finding frequency spectrum
Etc. feature, the voice vocal print feature that extraction obtains is sent to speech data acquisition module.
(6) speech data acquisition module 7 receives the voice signal that noise processed module is sent, may bag in this voice messaging
People's acoustic intelligence containing a small amount of noise and user；Speech data acquisition module 7 makes according to what the processing of vocal print processing module 4 was sent
The voice sound characteristic of user extracts the final speech data of user from above-mentioned voice messaging.
Based on the terminal shown in above-mentioned Fig. 3, the present embodiment additionally provides the tool of the noise-eliminating method of the voice signal shown in Fig. 4
Body flow chart, this method comprise the following steps：
Step S401, voice signal is received by the MIC of terminal, the voice signal includes：First MIC the first voice letter
Number and the 2nd MIC the second voice signal；
This step can be completed by above-mentioned MIC1 and MIC2.
Step S402, whether judge in the first voice signal and the second voice signal comprising the key message to prestore；If
It is to perform step S403, if not, terminating.
When including the key message to prestore in above-mentioned voice signal, illustrate that this needs to carry out de-noising pattern selection.
Step S403, the first voice signal and the second voice signal are replicated, be superimposed the first voice signal and second of duplication
Voice signal, generate speech superposition signal.
Step S402 and S403 can be completed by above-mentioned key message monitoring module 3.
Step S404, voice vocal print feature is extracted from speech superposition signal；This step can be complete by vocal print processing module 4
Into.
Step S405, compares the volume of the first voice signal and the second voice signal, and this step can be by above-mentioned ratio
Completed compared with device 5.
Step S406, de-noising pattern is chosen according to comparative result, the voice received believed according to the de-noising pattern of selection
Number carry out denoising Processing.
When the first MIC is main MIC, and the volume of the first voice signal subtracts the difference of the volume of the second voice signal and is more than
During given threshold, dual microphone noise reduction mode is chosen；Otherwise, steady-state noise tupe is chosen.
If the selection of de-noising pattern need not be carried out, can be handled according to relevant way, such as：It can be continuing with
The de-noising pattern for de-noising pattern or the terminal acquiescence that last time enables.
Step S406 can be completed by above-mentioned noise processed module 6.
The several line characteristic extraction procedures of above-mentioned steps S403 and step S404 behaviours, the process can be before step S405
Carry out, can also be carried out after step S405, for its order the present embodiment without limiting.
Step S407, according to voice vocal print feature, voice data extraction is carried out to the voice signal after denoising Processing, obtained
Final speech data.This step can be performed by speech data acquisition module 7.
The above method receives the first voice signal by the first MIC of terminal and the 2nd MIC receives the second voice signal, when
When needing to carry out the selection of de-noising pattern, compare the volume of the first voice signal and the second voice signal, and tie according to comparing
Fruit chooses de-noising pattern, and this mode is because terminal is configured with two kinds of de-noising patterns, and in the voice signal for comparing two MIC and receiving
Volume on the basis of choose de-noising pattern, the de-noising pattern of selection is more suitable for current MIC actual performance, therefore
The overall noise reduction of terminal is improved, improves user experience.Meanwhile after denoising Processing is carried out, according to extracting
Voice vocal print feature speech data is further extracted from the voice signal after de-noising, further improve voice signal
Definition.
It is simple for the concrete processing procedure of above-mentioned dual microphone noise reduction mode and steady-state noise tupe, the present embodiment
It is described as follows.
First, steady-state noise tupe
Terminal needs the voiceprint of prestored user and the ambient noise of place environment first, formed with being known with voice signal
It Ji He not carried with ambient noise set using identification set and ambient noise collection approach according to judgment mechanism Systematic selection
Useful human voice signal or the background sound taken out in current sound section is taken to improve the quality of useful voice signal, such as Fig. 5 institutes
The schematic flow sheet of the steady-state noise tupe shown, specifically includes following steps：
Step S501, text message is shown, and the voice data of text information is read aloud by microphone reception user.
Step S502, form identification set and background set.
Step S503, receive voice data.
Step S504, division sound section, the speech data obtained in set identification voice data is gathered using identification.
Step S505, judge whether to obtain speech data.If it is, perform step S507；If not, perform step
S506。
Step S506, the background sound in current sound section is removed by background set to obtain speech data.
Step S507, determine whether last sound section.If it is, terminate flow；If not, perform step
S508。
Step S508, using next sound section in the sound section of nearest one acquisition speech data as current sound
Section.Then step S505 is repeated to continue to determine whether to get speech data.
2nd, dual microphone noise reduction mode
The principle schematic of dual microphone noise reduction mode as shown in Figure 6, the main microphon of terminal is as the useful letter of pickup
Number i.e. voice, main MIC is typically placed in the low side of mobile phone；De-noising MIC pick up ambient noise, be typically placed in mobile phone top or
The back position of person's mobile phone upper part；By the dual microphone noise reduction algorithm inside system, based on energy subtraction method principle superposition
In original speech signal noise and the effect of raising call are eliminated to reach.Dual microphone noise reduction mode as shown in Figure 6
Principle schematic, illustrated therein is targeted voice signal acquisition module 601 and noise signal acquisition module 602 respectively believes sound
Number input to adaptive-filtering module 603 carry out adaptive-filtering processing, single channel denoising module 604 is entered data into after processing
Carry out denoising, the signal after final output denoising.
Different application scenarios are may relate in view of user's using terminal phonetic function, such as：Voice in phone
Function, the phonetic function in video, or phonetic function in Recording Process etc., the above method of the present embodiment can also wrap
Include：
If voice signal is the voice signal during telephone relation, final speech data is sent to call opposite end
Equipment；
If voice signal is the voice signal during video, the image in final speech data and video is sent to
The equipment of video opposite end；
If voice signal is the voice signal in Recording Process, final speech data is saved as into recorded audio.
Aforesaid way is by for different application scene, taking differentiation to handle final speech data, improving terminal
Process performance, the perfect function of terminal.
Embodiment three：
Fig. 7 shows the structured flowchart of the noise cancellation apparatus of the first voice signal provided in an embodiment of the present invention, the device
At least two MIC terminal is can be applied to, including：
Voice signal receiving module 701, for receiving voice signal by the MIC of terminal, wherein, voice signal includes：
First MIC the first voice signal and the 2nd MIC the second voice signal；
Volume comparison module 702, for when needing to carry out the selection of de-noising pattern, comparing the first voice signal and the second language
The volume of sound signal；
De-noising pattern chooses module 703, and for choosing de-noising pattern according to comparative result, de-noising pattern includes：Diamylose gram
Wind noise reduction mode and steady-state noise tupe；
Denoising Processing module 704, denoising Processing is carried out to the voice signal received for the de-noising pattern according to selection.
Fig. 8 shows the structured flowchart of the noise cancellation apparatus of second of voice signal provided in an embodiment of the present invention, Fig. 7's
On the basis of, in addition to：Key message judge module 801, for judging whether wrapped in the first voice signal and the second voice signal
Containing the key message to prestore；If it is, determine to need to carry out de-noising pattern selection.
Further, voice signal receiving module 701 is additionally operable to：When terminal starts voice signal receive capabilities, at end
The display interface at end shows re-reading prompt message, and re-reading prompt message includes the re-reading key message to prestore；Pass through the MIC of terminal
Receive voice signal corresponding to the key message to prestore.
Fig. 9 shows the structured flowchart of the noise cancellation apparatus of the third voice signal provided in an embodiment of the present invention, Fig. 7's
On the basis of, in addition to：De-noising pattern identification module 901, for when it is default judgement be identified as the first setting identification when, it is determined that need
Carry out de-noising pattern selection；When the de-noising pattern of selection is steady-state noise tupe, judgement is set to be identified as default
Second setting identification, and prompt the user with de-noising pattern.
Further, volume comparison module 702 is additionally operable to：Obtain the first energy and the second voice letter of the first voice signal
Number the second energy, the volume using the first energy as the first voice signal, the sound using the second energy as the second voice signal
Amount；Or the first intensity of the first voice signal and the second intensity of the second voice signal are obtained, using the first intensity as first
The volume of voice signal, the volume using the second intensity as the second voice signal.
Further, de-noising pattern is chosen module 703 and is additionally operable to：When the first MIC is main MIC, and the first voice signal
When the difference that volume subtracts the volume of the second voice signal is more than given threshold, dual microphone noise reduction mode is chosen；Otherwise, choose
Steady-state noise tupe.
Figure 10 shows the structured flowchart of the noise cancellation apparatus of the 4th kind of voice signal provided in an embodiment of the present invention, in Fig. 7
On the basis of, in addition to：Vocal print feature extraction module 1001, the first voice signal and the second voice signal are received for working as
When, replicate the first voice signal and the second voice signal；The first voice signal and the second voice signal replicated is superimposed, generates language
Sound superposed signal；Voice vocal print feature is extracted from speech superposition signal；Voice data extraction module 1002, for according to voice
Vocal print feature, voice data extraction is carried out to the voice signal after denoising Processing, obtains final speech data.
Figure 11 shows the structured flowchart of the noise cancellation apparatus of the 5th kind of voice signal provided in an embodiment of the present invention, in Figure 10
On the basis of, in addition to：Language data process module 1101, if being the voice letter during telephone relation for voice signal
Number, the equipment that final speech data is sent to call opposite end；If voice signal is the voice signal during video, will most
The equipment that image in whole speech data and video is sent to video opposite end；If voice signal is the voice letter in Recording Process
Number, final speech data is saved as into recorded audio.
The present embodiment additionally provides a kind of terminal, including memory and processor, and memory is used to store support processing
The program for the method that device above-described embodiment provides, processor are configurable for performing the program stored in memory.
The present embodiment additionally provides a kind of computer-readable storage medium, for saving as used in the device of above-described embodiment offer
Computer software instructions.
It should be noted that each embodiment in this specification is described by the way of progressive, each embodiment weight
Point explanation is all difference with other embodiment, between each embodiment identical similar part mutually referring to.
The noise cancellation apparatus and terminal for the voice signal that the embodiment of the present invention is provided, its realization principle and caused technique effect and foregoing
Embodiment of the method is identical, and to briefly describe, device embodiment part does not refer to part, refers to corresponding in preceding method embodiment
Content.
In several embodiments provided herein, it should be understood that disclosed apparatus and method, can also pass through
Other modes are realized.Device embodiment described above is only schematical, for example, flow chart and block diagram in accompanying drawing
Show the device of multiple embodiments according to the present invention, method and computer program product architectural framework in the cards,
Function and operation.At this point, each square frame in flow chart or block diagram can represent the one of a module, program segment or code
Part, a part for the module, program segment or code include one or more and are used to realize holding for defined logic function
Row instruction.It should also be noted that at some as in the implementation replaced, the function that is marked in square frame can also with different from
The order marked in accompanying drawing occurs.For example, two continuous square frames can essentially perform substantially in parallel, they are sometimes
It can perform in the opposite order, this is depending on involved function.It is it is also noted that every in block diagram and/or flow chart
The combination of individual square frame and block diagram and/or the square frame in flow chart, function or the special base of action as defined in performing can be used
Realize, or can be realized with the combination of specialized hardware and computer instruction in the system of hardware.
In addition, each functional module or unit in each embodiment of the present invention can integrate and to form an independence
Part or modules individualism, can also two or more modules be integrated to form an independent portion
Point.
If the function is realized in the form of software function module and is used as independent production marketing or in use, can be with
It is stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially in other words
The part to be contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are causing a computer equipment (can be
People's computer, server, or network equipment etc.) perform all or part of step of each embodiment methods described of the present invention.
And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.
It should be noted that herein, such as first and second or the like relational terms are used merely to a reality
Body or operation make a distinction with another entity or operation, and not necessarily require or imply and deposited between these entities or operation
In any this actual relation or order, can not be interpreted as indicating or implying relative importance.Moreover, term " bag
Include ", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion so that the mistake including a series of elements
Journey, method, article or equipment not only include those key elements, but also the other element including being not expressly set out, either
Also include for this process, method, article or the intrinsic key element of equipment.In the absence of more restrictions, by sentence
The key element that "including a ..." limits, it is not excluded that also deposited in the process including the key element, method, article or equipment
In other identical element.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies
Change, equivalent substitution, improvement etc., should be included in the scope of the protection.It should be noted that：Similar label and letter exists
Similar terms is represented in following accompanying drawing, therefore, once being defined in a certain Xiang Yi accompanying drawing, is then not required in subsequent accompanying drawing
It is further defined and explained.
The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all be contained
Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.
The embodiment of the present invention further discloses：
A1. a kind of noise-eliminating method of voice signal, applied at least two MIC terminal, methods described includes：
Voice signal is received by the MIC of the terminal, wherein, the voice signal includes：First MIC the first voice
Second voice signal of signal and the 2nd MIC；
When needing to carry out the selection of de-noising pattern, the volume of first voice signal and second voice signal
Size；
De-noising pattern is chosen according to comparative result, the de-noising pattern includes：Dual microphone noise reduction mode and steady-state noise
Tupe；
Denoising Processing is carried out to the voice signal received according to the de-noising pattern of selection.
A2. the method according to A1, methods described also include：
Whether judge in first voice signal and second voice signal comprising the key message to prestore；
If it is, determine to need to carry out de-noising pattern selection.
A3. the method according to A2, the step of MIC by the terminal receives voice signal, include：
When the terminal starts voice signal receive capabilities, re-reading prompting letter is shown in the display interface of the terminal
Breath, the re-reading prompt message include the re-reading key message to prestore；
By the MIC of the terminal receive described in voice signal corresponding to the key message that prestores.
A4. the method according to A1, methods described also include：
When default judgement is identified as the first setting identification, it is determined that needing to carry out de-noising pattern selection；
When the de-noising pattern of selection is steady-state noise tupe, the judgement is set to be identified as default second setting
Mark, and prompt the user with the de-noising pattern.
A5. the volume of the method according to A1, first voice signal and second voice signal
The step of size, includes：
The first energy of first voice signal and the second energy of second voice signal are obtained, by described first
Volume of the energy as first voice signal, the volume using second energy as second voice signal；Or
The first intensity of first voice signal and the second intensity of second voice signal are obtained, by described first
Volume of the intensity as first voice signal, the volume using second intensity as second voice signal.
A6. the method according to A1, described the step of choosing de-noising pattern according to comparative result, include：
When the first MIC is main MIC, and the volume of first voice signal subtracts the sound of second voice signal
When the difference of amount is more than given threshold, dual microphone noise reduction mode is chosen；Otherwise, steady-state noise tupe is chosen.
A7. the method according to A1, methods described also include：When receiving first voice signal and described second
During voice signal, first voice signal and second voice signal are replicated；It is superimposed first voice signal replicated
With second voice signal, speech superposition signal is generated；Voice vocal print feature is extracted from the speech superposition signal；
The step of de-noising pattern according to selection carries out denoising Processing to the voice signal that receives it
Afterwards, methods described also includes：
According to the voice vocal print feature, voice data extraction is carried out to the voice signal after denoising Processing, obtained
Final speech data.
A8. the method according to A7, methods described also include：
If the voice signal is the voice signal during telephone relation, the final speech data is sent to logical
The equipment for talking about opposite end；
If the voice signal is the voice signal during video, by the final speech data and the video
The image equipment that is sent to video opposite end；
If the voice signal is the voice signal in Recording Process, the final speech data is saved as into recording sound
Frequently.
B9. a kind of noise cancellation apparatus of voice signal, applied at least two MIC terminal, described device includes：
Voice signal receiving module, for receiving voice signal by the MIC of the terminal, wherein, the voice signal
Including：First MIC the first voice signal and the 2nd MIC the second voice signal；
Volume comparison module, for when needing to carry out de-noising pattern to choose, first voice signal and described
The volume of second voice signal；
De-noising pattern chooses module, and for choosing de-noising pattern according to comparative result, the de-noising pattern includes：Diamylose gram
Wind noise reduction mode and steady-state noise tupe；
Denoising Processing module, de-noising is carried out to the voice signal received for the de-noising pattern according to selection
Processing.
B10. the device according to B9, described device also include：
Key message judge module, for judging whether included in first voice signal and second voice signal
The key message to prestore；If it is, determine to need to carry out de-noising pattern selection.
B11. the device according to B10, the voice signal receiving module are additionally operable to：
When the terminal starts voice signal receive capabilities, re-reading prompting letter is shown in the display interface of the terminal
Breath, the re-reading prompt message include the re-reading key message to prestore；
By the MIC of the terminal receive described in voice signal corresponding to the key message that prestores.
B12. the device according to B9, described device also include：
De-noising pattern identification module, for when it is default judgement be identified as the first setting identification when, it is determined that need disappeared
Pattern of making an uproar is chosen；When the de-noising pattern of selection is steady-state noise tupe, the judgement is set to be identified as default second
Setting identification, and prompt the user with the de-noising pattern.
B13. the device according to B9, the volume comparison module are additionally operable to：
The first energy of first voice signal and the second energy of second voice signal are obtained, by described first
Volume of the energy as first voice signal, the volume using second energy as second voice signal；Or
The first intensity of first voice signal and the second intensity of second voice signal are obtained, by described first
Volume of the intensity as first voice signal, the volume using second intensity as second voice signal.
B14. the device according to B9, the de-noising pattern are chosen module and are additionally operable to：When the first MIC is main MIC,
And the volume of first voice signal is chosen double when subtracting the difference of the volume of second voice signal and being more than given threshold
Microphone denoising pattern；Otherwise, steady-state noise tupe is chosen.
B15. the device according to B9, described device also include：
Vocal print feature extraction module, for when receiving first voice signal and second voice signal, answering
Make first voice signal and second voice signal；It is superimposed first voice signal replicated and second voice
Signal, generate speech superposition signal；Voice vocal print feature is extracted from the speech superposition signal；
Voice data extraction module, for according to the voice vocal print feature, to the voice signal after denoising Processing
Voice data extraction is carried out, obtains final speech data.
B16. the device according to B15, described device also include：
Language data process module, if being the voice signal during telephone relation for the voice signal, by institute
State the equipment that final speech data is sent to call opposite end；, will if the voice signal is the voice signal during video
The equipment that image in the final speech data and the video is sent to video opposite end；If the speech data is recording
During voice signal, the final speech data is saved as into recorded audio.
C17. a kind of mobile terminal, including memory and processor, the memory, which is used to store, supports processor to hold
The program of any one of row A1 to A8 methods described, the processor are configurable for performing the journey stored in the memory
Sequence.
D18. a kind of computer-readable storage medium is soft for saving as computer used in any one of B9 to B16 described device
Part instructs.
Claims (10)
1. a kind of noise-eliminating method of voice signal, it is characterised in that applied at least two MIC terminal, methods described includes：
Voice signal is received by the MIC of the terminal, wherein, the voice signal includes：First MIC the first voice signal
With the 2nd MIC the second voice signal；
When needing to carry out de-noising pattern to choose, first voice signal and second voice signal give great volume
It is small；
De-noising pattern is chosen according to comparative result, the de-noising pattern includes：Dual microphone noise reduction mode and steady-state noise processing
Pattern；
Denoising Processing is carried out to the voice signal received according to the de-noising pattern of selection.
2. according to the method for claim 1, it is characterised in that methods described also includes：
Whether judge in first voice signal and second voice signal comprising the key message to prestore；
If it is, determine to need to carry out de-noising pattern selection.
3. according to the method for claim 2, it is characterised in that the MIC by the terminal receives voice signal
Step includes：
When the terminal starts voice signal receive capabilities, re-reading prompt message, institute are shown in the display interface of the terminal
Stating re-reading prompt message includes the re-reading key message to prestore；
By the MIC of the terminal receive described in voice signal corresponding to the key message that prestores.
4. according to the method for claim 1, it is characterised in that methods described also includes：
When default judgement is identified as the first setting identification, it is determined that needing to carry out de-noising pattern selection；
When the de-noising pattern of selection is steady-state noise tupe, the judgement is set to be identified as default second setting mark
Know, and prompt the user with the de-noising pattern.
5. according to the method for claim 1, it is characterised in that first voice signal and second language
The step of volume of sound signal, includes：
The first energy of first voice signal and the second energy of second voice signal are obtained, by first energy
As the volume of first voice signal, the volume using second energy as second voice signal；Or
The first intensity of first voice signal and the second intensity of second voice signal are obtained, by first intensity
As the volume of first voice signal, the volume using second intensity as second voice signal.
6. according to the method for claim 1, it is characterised in that described the step of choosing de-noising pattern according to comparative result wraps
Include：
When the first MIC is main MIC, and the volume of first voice signal subtracts the volume of second voice signal
When difference is more than given threshold, dual microphone noise reduction mode is chosen；Otherwise, steady-state noise tupe is chosen.
7. according to the method for claim 1, it is characterised in that methods described also includes：When receiving first voice
When signal and second voice signal, first voice signal and second voice signal are replicated；The institute that superposition replicates
The first voice signal and second voice signal are stated, generates speech superposition signal；People is extracted from the speech superposition signal
Sound vocal print feature；
After the step of de-noising pattern according to selection carries out denoising Processing to the voice signal received, institute
Stating method also includes：
According to the voice vocal print feature, voice data extraction is carried out to the voice signal after denoising Processing, obtained final
Speech data.
8. according to the method for claim 7, it is characterised in that methods described also includes：
If the voice signal is the voice signal during telephone relation, the final speech data is sent to call pair
The equipment at end；
If the voice signal is the voice signal during video, by the figure in the final speech data and the video
Equipment as being sent to video opposite end；
If the voice signal is the voice signal in Recording Process, the final speech data is saved as into recorded audio.
9. a kind of noise cancellation apparatus of voice signal, it is characterised in that applied at least two MIC terminal, described device includes：
Voice signal receiving module, for receiving voice signal by the MIC of the terminal, wherein, the voice signal includes：
First MIC the first voice signal and the 2nd MIC the second voice signal；
Volume comparison module, for when needing to carry out de-noising pattern to choose, first voice signal and described second
The volume of voice signal；
De-noising pattern chooses module, and for choosing de-noising pattern according to comparative result, the de-noising pattern includes：Dual microphone drops
Pattern of making an uproar and steady-state noise tupe；
Denoising Processing module, the voice signal received is carried out at de-noising for the de-noising pattern according to selection
Reason.
10. a kind of mobile terminal, it is characterised in that including memory and processor, the memory is used to store at support
The program of any one of device perform claim requirement 1 to 8 methods described is managed, the processor is configurable for performing the memory
The program of middle storage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710810085.3A CN107369441A (en) | 2017-09-08 | 2017-09-08 | Noise-eliminating method, device and the terminal of voice signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710810085.3A CN107369441A (en) | 2017-09-08 | 2017-09-08 | Noise-eliminating method, device and the terminal of voice signal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107369441A true CN107369441A (en) | 2017-11-21 |
Family
ID=60312009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710810085.3A Pending CN107369441A (en) | 2017-09-08 | 2017-09-08 | Noise-eliminating method, device and the terminal of voice signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107369441A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108234766A (en) * | 2017-12-29 | 2018-06-29 | 努比亚技术有限公司 | Microphone switching method, mobile terminal and computer readable storage medium |
| CN108305637A (en) * | 2018-01-23 | 2018-07-20 | 广东欧珀移动通信有限公司 | Earphone method of speech processing, terminal device and storage medium |
| CN108712551A (en) * | 2018-06-04 | 2018-10-26 | 联想(北京)有限公司 | A kind of call control method and electronic equipment |
| CN110060689A (en) * | 2019-04-10 | 2019-07-26 | 南京启诺信息技术有限公司 | A kind of intelligent wearable device and its application method with the identification of feature voice |
| CN111970410A (en) * | 2020-08-26 | 2020-11-20 | 展讯通信（上海）有限公司 | Echo cancellation method and device, storage medium and terminal |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102801861A (en) * | 2012-08-07 | 2012-11-28 | 歌尔声学股份有限公司 | Voice enhancing method and device applied to cell phone |
| CN103310794A (en) * | 2012-03-11 | 2013-09-18 | 闻泰通讯股份有限公司 | Method filtering conversation noise and device filtering conversation noise |
| CN104335600A (en) * | 2013-02-25 | 2015-02-04 | 展讯通信（上海）有限公司 | Detecting and switching between noise reduction modes in multi-microphone mobile devices |
| CN104754430A (en) * | 2013-12-30 | 2015-07-01 | 重庆重邮信科通信技术有限公司 | Noise reduction device and method for terminal microphone |
| CN104811559A (en) * | 2015-05-05 | 2015-07-29 | 上海青橙实业有限公司 | Noise reduction method, communication method and mobile terminal |
| CN106302906A (en) * | 2016-09-15 | 2017-01-04 | 上海斐讯数据通信技术有限公司 | A kind of diamylose gram noise-reduction method and device, a kind of mobile terminal |
| CN106782589A (en) * | 2016-12-12 | 2017-05-31 | 奇酷互联网络科技（深圳）有限公司 | Mobile terminal and its pronunciation inputting method and device |
-
2017
- 2017-09-08 CN CN201710810085.3A patent/CN107369441A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103310794A (en) * | 2012-03-11 | 2013-09-18 | 闻泰通讯股份有限公司 | Method filtering conversation noise and device filtering conversation noise |
| CN102801861A (en) * | 2012-08-07 | 2012-11-28 | 歌尔声学股份有限公司 | Voice enhancing method and device applied to cell phone |
| CN104335600A (en) * | 2013-02-25 | 2015-02-04 | 展讯通信（上海）有限公司 | Detecting and switching between noise reduction modes in multi-microphone mobile devices |
| CN104754430A (en) * | 2013-12-30 | 2015-07-01 | 重庆重邮信科通信技术有限公司 | Noise reduction device and method for terminal microphone |
| CN104811559A (en) * | 2015-05-05 | 2015-07-29 | 上海青橙实业有限公司 | Noise reduction method, communication method and mobile terminal |
| CN106302906A (en) * | 2016-09-15 | 2017-01-04 | 上海斐讯数据通信技术有限公司 | A kind of diamylose gram noise-reduction method and device, a kind of mobile terminal |
| CN106782589A (en) * | 2016-12-12 | 2017-05-31 | 奇酷互联网络科技（深圳）有限公司 | Mobile terminal and its pronunciation inputting method and device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108234766A (en) * | 2017-12-29 | 2018-06-29 | 努比亚技术有限公司 | Microphone switching method, mobile terminal and computer readable storage medium |
| CN108305637A (en) * | 2018-01-23 | 2018-07-20 | 广东欧珀移动通信有限公司 | Earphone method of speech processing, terminal device and storage medium |
| CN108305637B (en) * | 2018-01-23 | 2021-04-06 | Oppo广东移动通信有限公司 | Earphone voice processing method, terminal equipment and storage medium |
| CN108712551A (en) * | 2018-06-04 | 2018-10-26 | 联想(北京)有限公司 | A kind of call control method and electronic equipment |
| CN110060689A (en) * | 2019-04-10 | 2019-07-26 | 南京启诺信息技术有限公司 | A kind of intelligent wearable device and its application method with the identification of feature voice |
| CN111970410A (en) * | 2020-08-26 | 2020-11-20 | 展讯通信（上海）有限公司 | Echo cancellation method and device, storage medium and terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107369441A (en) | Noise-eliminating method, device and the terminal of voice signal | |
| CN107566658A (en) | Call method, device, storage medium and mobile terminal | |
| US8600743B2 (en) | Noise profile determination for voice-related feature | |
| CN102638612B (en) | Method and device for ending conversation | |
| EP2821992A1 (en) | Method for updating voiceprint feature model and terminal | |
| CN106920559B (en) | Voice communication optimization method and device and call terminal | |
| CN104883437B (en) | The method and system of speech analysis adjustment reminding sound volume based on environment | |
| CN107919138B (en) | Emotion processing method in voice and mobile terminal | |
| CN103491251A (en) | Method and terminal for monitoring user calls | |
| CN107231473A (en) | A kind of audio output regulation and control method, equipment and computer-readable recording medium | |
| CN106534532A (en) | Call quality improving method and terminal | |
| CN107331405A (en) | A kind of voice information processing method and server | |
| CN108111986A (en) | A kind of conversion method of notification message, terminal and computer readable storage medium | |
| CN108766453A (en) | Voice de-noising method, device, readable storage medium storing program for executing and mobile terminal | |
| CN104202485B (en) | A kind of safety call method, device and mobile terminal | |
| CN108391190A (en) | A kind of noise-reduction method, earphone and computer readable storage medium | |
| CN112259112A (en) | Echo cancellation method combining voiceprint recognition and deep learning | |
| CN111491236A (en) | Active noise reduction earphone, awakening method and device thereof and readable storage medium | |
| CN104821985B (en) | A kind of call control method and device | |
| CN109256151A (en) | Call voice regulates and controls method, apparatus, mobile terminal and readable storage medium storing program for executing | |
| WO2022068694A1 (en) | Electronic device and wake-up method thereof | |
| CN107659864A (en) | A kind of noise-reduction method and terminal | |
| CN108712551A (en) | A kind of call control method and electronic equipment | |
| CN108924330A (en) | Information processing method, device and electronic equipment | |
| CN104735249A (en) | Information processing method and electronic equipment |
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: 20171121 |
|
| RJ01 | Rejection of invention patent application after publication |