CN106453762B - The processing method and system that voice is uttered long and high-pitched sounds in audio system - Google Patents
The processing method and system that voice is uttered long and high-pitched sounds in audio system Download PDFInfo
- Publication number
- CN106453762B CN106453762B CN201610946477.8A CN201610946477A CN106453762B CN 106453762 B CN106453762 B CN 106453762B CN 201610946477 A CN201610946477 A CN 201610946477A CN 106453762 B CN106453762 B CN 106453762B
- Authority
- CN
- China
- Prior art keywords
- voice
- signal
- pitched sounds
- frequency domain
- sound
- 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.)
- Active
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 71
- 238000004458 analytical method Methods 0.000 claims abstract description 6
- 230000005540 biological transmission Effects 0.000 claims abstract description 4
- 238000001514 detection method Methods 0.000 claims description 32
- 238000000034 method Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 239000012141 concentrate Substances 0.000 claims 1
- 239000004744 fabric Substances 0.000 claims 1
- 230000000694 effects Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 230000005236 sound signal Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000006837 decompression Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/02—Constructional features of telephone sets
- H04M1/19—Arrangements of transmitters, receivers, or complete sets to prevent eavesdropping, to attenuate local noise or to prevent undesired transmission; Mouthpieces or receivers specially adapted therefor
-
- 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)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The invention discloses processing method and system that voice in audio system is uttered long and high-pitched sounds, this programme by being handled in real time in frequency domain voice signal, analysis check signal whether the useful signal containing voice sound；When not detecting human voice signal, illustrates that background may include more noise or utter long and high-pitched sounds, convert back time-domain transmission after frequency domain clear signal.This programme is only processed when the voice for having noise or uttering long and high-pitched sounds, and will not influence normal talking quality；After transmitting terminal recording, addition processing before the broadcasting of receiving end, the hardware or software design of existed system are not influenced.
Description
Technical field
The present invention relates to the processing techniques that voice in audio system processing technique more particularly to audio system is uttered long and high-pitched sounds.
Background technique
With the development of digital network, there is numerous mobile phone speech intercom systems: can be very by cell phone software
Easily realize traditional intercom.Transmitting terminal mobile phone is obtained voice signal and is transmitted to by data network and answered by Mike
End answers end by loudspeaker and plays the voice signal received, and such a basic voice inter-speaking system is formed.But
It is that in actual application, if conversed indoors or closely, the loudspeaker due to answering end is constantly made a sound, together
When also received by the Mike of transmitting terminal, constantly circulation produce the self-excitation of sound, there is ear-piercing sound of uttering long and high-pitched sounds.This sound of uttering long and high-pitched sounds
The experience effect of system is destroyed very serious, it is therefore desirable to a kind of technology that can effectively filter out sound of uttering long and high-pitched sounds.
In existing technology, the method uttered long and high-pitched sounds of processing voice has very much, such as using sef-adapting filter reduction gain or
Trapper；Signal center frequency is such as judged whether in the frequency range of generation of uttering long and high-pitched sounds, if it find that then swashing in occurrence frequency of uttering long and high-pitched sounds
The specific frequency of filters filter living；Or frequency of specifically uttering long and high-pitched sounds directly is filtered using trapper.
Traditional audio system utter long and high-pitched sounds processing technique mostly be to be processed in signal gain, may utter long and high-pitched sounds when detecting
Afterwards, gain is reduced to audio signal, avoids voice system circuit from positive feedback self-excitation formation occur and utters long and high-pitched sounds.Reduce the method meeting of gain
Normal voice is influenced simultaneously, so that receiving end voice signal is become smaller, and do not filtered completely due to uttering long and high-pitched sounds, it would still be possible to will continue to
Self-excitation generation is more serious to utter long and high-pitched sounds, and treatment effect is unsatisfactory.
Directly certain can be effectively removed using the method that trap filters fall specific frequency to utter long and high-pitched sounds, but with audio
The continuous feedback of system circuit, sound of uttering long and high-pitched sounds can constantly change, i.e., center frequency constantly shifts.And use adaptive trapper
Processing, then may filter out normal acoustic information, effect is still undesirable.
Summary of the invention
It utters long and high-pitched sounds the problems of processing technique for existing audio system, needs new audio system to utter long and high-pitched sounds and handle skill
Art, to improve treatment effect.
The processing method uttered long and high-pitched sounds as a result, technical problem to be solved by the invention is to provide voice in a kind of audio system and
System, to improve treatment effect of uttering long and high-pitched sounds.
In order to solve the above technical problem, the present invention provides audio system in the processing method uttered long and high-pitched sounds of voice, pass through
Detection voice sound and utter long and high-pitched sounds sound frequency domain feature, it is determined whether sound of uttering long and high-pitched sounds, and utter long and high-pitched sounds sound occur when remove its own
Frequency domain components.
Preferably, uniformly whether the processing method be distributed by detecting voice signal in voice frequency domain, is uniformly then
Otherwise normal speech signals containing voice sound are background noise or sound of uttering long and high-pitched sounds.
Preferably, the processing method specifically includes: is handled in real time in frequency domain voice signal, analysis checks voice
Signal whether the normal speech signals containing voice sound；When not detecting human voice signal, it is determined as background noise or utters long and high-pitched sounds
Signal converts back time-domain transmission after frequency domain removes the signal.
Preferably, the processing method determines letter by checking strength characteristic of the voice signal in voice frequency domain
Whether uniformly number it is distributed in voice frequency domain.
Preferably, the processing method is directed to voice signal every band frequency component in voice frequency domain, traverses each section of frequency
The energy intensity of rate component, when energy it is powerful in the intensity measured in advance when counting, when count reach the effective people measured in advance
When sound counts, determine that voice signal is normal speech signals, it is without any processing；When count do not reach measure in advance it is effective
When voice counts, illustrate being unevenly distributed for voice signal, be determined as signal of uttering long and high-pitched sounds, frequency-region signal is reset and handles and is converted back into
Time domain eliminates signal of uttering long and high-pitched sounds.
Preferably, the position of voice frequency domain is determined in the processing method by following formula:
Low=(int) (size*1000f*2/rate),
Wherein, low is the position of people's acoustic low-pass, and rate is input audio specimen sample rate, and size is sample size.
In order to solve the above technical problem, the present invention provides audio system in the processing system uttered long and high-pitched sounds of voice include:
FFT module, the FFT module are used to audio sample progress Fast Fourier Transform going to frequency domain；
Sound detection module, the sound detection module carry out sound detection to the signal for transforming to frequency domain, determine signal
Whether to be uniformly distributed in voice frequency domain；
Sound processing module, the sound processing module eliminate signal according to the testing result of sound detection module
It utters long and high-pitched sounds processing；
IFFT module, the IFFT module carry out inverse-Fourier transform for signal to the signal handled through sound processing module
It is converted back to time domain.
Preferably, the sound detection module is by checking strength characteristic of the voice signal in voice frequency domain, really
Determine whether uniformly signal is distributed in voice frequency domain.
Preferably, the sound detection module check voice signal every band frequency component in voice frequency domain, and by every section
The energy intensity of frequency component and the intensity contrast measured in advance are counted when being higher than preparatory measurement intensity；When counting reaches pre-
When the effective voice first measured counts, determine that signal is evenly distributed in voice frequency domain；When what counting not up to measured in advance has
When imitating voice counting, determine that signal is distributed concentration in voice frequency domain.
Preferably, the sound processing module is not processed the signal being evenly distributed in voice frequency domain；To in voice
Signal unevenly distributed in frequency domain removes its all frequency domain components.
It preferably, further include initialization module in the processing system, the initialization module is to determine voice frequency domain
Position.
It is opposite with existing processing method, audio system provided by the invention processing scheme of uttering long and high-pitched sounds has the advantages that
1) audio for not including effective sound is filtered completely, avoids noise and uttering long and high-pitched sounds and is passed again in phonological loop
It passs；
2) any processing will not be done to normal audio signal, it is ensured that normal speech signals gain is constant and without appointing
What frequency and energy loss；
3) implement simply, it is easy to operate, it does not need any hardware and software and changes, can quickly be deployed in existing
Voice system；
4) test has ideal utter long and high-pitched sounds and noise treatment effect in true practical application；
5) it is only processed when the voice for having noise or uttering long and high-pitched sounds, will not influence normal talking quality；
6) this programme does not influence the hardware or soft of existed system after transmitting terminal recording, addition processing before the broadcasting of receiving end
Part design.
Detailed description of the invention
The present invention is further illustrated below in conjunction with the drawings and specific embodiments.
Fig. 1 is to utter long and high-pitched sounds to handle model schematic in present example；
Fig. 2 is Processing Algorithm flow chart of uttering long and high-pitched sounds in present example；
Fig. 3 is to increase audio system model of uttering long and high-pitched sounds that treated in present example；
Fig. 4 is that a untreated radio voice system is uttered long and high-pitched sounds the spectrum signature figure of audio sample；
Fig. 5 is a frequency spectrum of the radio voice system without audio sample of uttering long and high-pitched sounds by processing scheme processing of the invention of uttering long and high-pitched sounds
Characteristic pattern.
Specific embodiment
In order to be easy to understand the technical means, the creative features, the aims and the efficiencies achieved by the present invention, tie below
Conjunction is specifically illustrating, and the present invention is further explained.
Based on a large amount of experiment and creative work, it is found by the applicant that the sound of people has apparent feature in frequency domain: small
The band (as shown in Figure 5) being evenly distributed is shown as in the low frequency part of 1000hz；And obvious spy of the signal in frequency domain that utter long and high-pitched sounds
Sign: integrated distribution (as shown in Figure 4) in a certain band frequency；It in the case, although can using traditional processing method of uttering long and high-pitched sounds
To inhibit the signal of uttering long and high-pitched sounds of specific frequency, but in real-time audio system, sound of uttering long and high-pitched sounds may become because inhibiting processing
Change, centre frequency of uttering long and high-pitched sounds can be shifted constantly, and at this moment traditional processing method is difficult to reach ideal effect.
For this purpose, the present invention by detection voice sound and utter long and high-pitched sounds sound frequency domain feature, to determine whether sound of uttering long and high-pitched sounds, when
Sound of uttering long and high-pitched sounds removes all frequency domain components when occurring, and thus eliminates sound of uttering long and high-pitched sounds.
For wherein, whether voice sound, can uniform in low frequency distribution by detection voice in the detection of the feature of frequency domain,
Even is then the normal speech signals containing voice sound, is otherwise background noise or sound of uttering long and high-pitched sounds.
Accordingly scheme to voice in audio system utter long and high-pitched sounds processing process it is as follows:
Voice signal is handled in real time in frequency domain, checks whether voice signal contains voice sound by phonetic analysis
Normal speech signals；
When not detecting human voice signal, illustrate that background may include more noise or utter long and high-pitched sounds, removing in frequency domain should
Time-domain transmission is converted back after signal.
Whether check to be distributed by detection voice in low frequency containing voice sound signal in voice signal in this programme is
No uniform come what is carried out, detailed process is as follows:
1) it to audio sample to be detected, carries out Fast Fourier Transform and goes to frequency domain；
2) audio sample signal each frequency point in voice frequency range 50hz-1000hz frequency domain after centralized processing transformation
Amount checks that whether uniformly the audio sample signal is distributed in voice frequency domain, exists here especially by inspection and analysis voice signal
Uniformly whether the strength characteristic in voice frequency domain be distributed with this to determine signal in voice frequency domain, and process is as follows:
Detect the energy intensity px of each frequency component, and by the energy intensity px of each frequency component in advance measure
Intensity p0 comparison, counts when being higher than preparatory measurement intensity p0；
When counting, which reaches the effective voice measured in advance, counts c0, then sound detection is triggered, illustrates that the audio sample is believed
It number is evenly distributed in voice frequency domain, it is without any processing to signal for the normal speech signals containing voice sound；When count not
When reaching the effective voice counting c0 measured in advance, then illustrate that the audio sample signal is unevenly distributed (i.e. in voice frequency domain
For integrated distribution), which is background noise or sound of uttering long and high-pitched sounds, and is reset to all frequency domain components of audio sample signal；
3) by treated, signal converts back time domain, is transmitted or is played by existed system.
From the foregoing, it will be observed that processing scheme of originally uttering long and high-pitched sounds only is processed when the voice for having noise or uttering long and high-pitched sounds, positive normal open will not influence
Talk about quality.
Come illustratively below by way of a specific application example.
It utters long and high-pitched sounds processing in this example for voice in above-mentioned audio system, designs a kind of voice for audio system
It utters long and high-pitched sounds processing system.
Referring to Fig. 1 which shows voice is uttered long and high-pitched sounds the composition schematic diagram of processing system in this example.
As seen from the figure, this voice utters long and high-pitched sounds processing system 100 mainly including FFT module 110, sound detection module 120, sound
Processing module 130 and IFFT module 140.
Wherein, FFT module 110 go to frequency domain for audio sample to be processed to be carried out Fast Fourier Transform.
Sound detection module 120, it is to be processed with determination for carrying out sound detection analysis to the signal for transforming to frequency domain
Uniformly whether signal be distributed in voice frequency domain.
The sound detection module 120 is by checking strength characteristic of the voice signal to be processed in voice frequency domain, really
Determine whether uniformly signal is distributed in voice frequency domain.
Specifically, this sound detection module 120 pass through to transformed audio sample (data []) it is all in voice frequency
Frequency component in domain is traversed, and whether more each frequency component values (data [i]) are greater than preset threshold value
(power), meet condition hour counter (count) plus one；Finally check whether counter (count) is greater than preset measured value
(C)；If more than then judging that signal is evenly distributed in voice frequency domain；If being less than, judge that signal is distributed collection in voice frequency domain
In.
Here preset threshold value (power) is the intensity measured in advance；And preset measured value (C) measures in advance
Effective voice counts.
Sound processing module 130 carries out signal to eliminate the place that utters long and high-pitched sounds for the testing result according to sound detection module 120
Reason.
If it is to be evenly distributed that the sound processing module 130, which receives 120 testing result of sound detection module, it is determined that the sound
Frequency sample (data []) is normal voice signal, without any processing to its；If the sound processing module 130 receives sound
120 testing result of sound detection module is that distribution is concentrated, it is determined that the audio sample (data []) is background noise or letter of uttering long and high-pitched sounds
Number, operation is zeroed out to audio sample (data []) all frequency domain components.
IFFT module 140, for through sound processing module 130 handle (including it is no processing or to all frequency domain components
Be zeroed out) signal carry out inverse-Fourier transform convert signals back to time domain and export.
It utters long and high-pitched sounds the effect of the processing of processing system 100, is also further added in the system just in order to further increase this voice
Beginningization module, for determining the position of voice frequency domain in real time.The initialization module is according to input audio specimen sample rate and sampling
Thus size realizes real-time determination determining the position (low) of sound detection upper frequency limit (1000hz) i.e. people's acoustic low-pass
The position of voice frequency domain.
Accordingly, which realizes especially by following formulation schema:
low=(int)(size*1000f*2/rate)；
Wherein, rate is input audio specimen sample rate, and size is sample size.
Referring to fig. 2 which shows this voice processing system 100 of uttering long and high-pitched sounds utter long and high-pitched sounds the flow chart of processing.
As seen from the figure, this voice utter long and high-pitched sounds process of processing of processing system 100 of uttering long and high-pitched sounds includes system initialization and utters long and high-pitched sounds
Handle two parts.
Initialization:
Input audio specimen sample rate (rate) and sample size (size), initialization module pass through formula low=(int)
(size*1000f*2/rate) position (low) of sound detection upper frequency limit (1000hz) i.e. people's acoustic low-pass is calculated.
It utters long and high-pitched sounds processing:
Input audio sample (data []) arrives FFT module, and FFT module carries out Fast Fourier Transform to sample and goes to frequency
Domain, and it is input to sound detection module；
All frequencies that are less than detection upper frequency limit (low) of the sound detection module to transformed audio sample (data [])
Rate component is traversed, and compares whether component value (data [i]) is greater than preset threshold value (power), meets condition hour counter
(count) add one；
Check whether counter (count) is greater than preset measured value (C), explanation is normal voice letter when meeting condition
Number, without any processing, otherwise explanation is background noise or signal of uttering long and high-pitched sounds, to audio sample (data []) clear operation；
Inverse-Fourier transform is done to audio sample (data []), return to time domain and exports audio sample (data []).
Referring to Fig. 3 which shows there is the sound of processing function of uttering long and high-pitched sounds based on what above-mentioned voice uttered long and high-pitched sounds the composition of processing system 100
Display system.The entirety audio system model is broadly divided into two parts: transmitting terminal 200 and receiving end 300.
Wherein, transmitting terminal main body 200 includes that microphone and processing unit of uttering long and high-pitched sounds (are uttered long and high-pitched sounds 100 structure of processing system by voice
At), then receiving end is transferred to by compressed encoding and physical link；
Receiving end main body 300 includes loudspeaker and processing unit of uttering long and high-pitched sounds (being made of voice processing system 100 of uttering long and high-pitched sounds), is passed through
After physical link receives the voice signal that transmitting terminal is sent, carries out decompression and decoding process and pass to processing unit of uttering long and high-pitched sounds.
The whole audio system constituted accordingly can be handled the sound of uttering long and high-pitched sounds generated in audio system, be greatly improved logical
Talk about quality.
Here audio system is the audio system of the various sound processing that utter long and high-pitched sounds, as an example such as mobile phone communication pair
Say system etc..Below by increase voice utter long and high-pitched sounds processing mobile phone communication intercom system for, illustratively its carry out voice utter long and high-pitched sounds
The process of processing.
1, transmitting terminal: transmitting terminal gets voice PCM signal from mobile phone Mike, does FFT transform to the signal got and turns
The frequency domain of change checks signal in the strength characteristic of voice frequency domain 50hz-1000hz；The energy for traversing each band frequency component is strong
Degree increases when signal strength is greater than preset threshold value and counts；When counter does not reach preset threshold value, illustrate signal
It is unevenly distributed, can be determined that signal of uttering long and high-pitched sounds；Frequency-region signal is reset and handles and is converted back into time domain to eliminate signal of uttering long and high-pitched sounds, is located
Signal after reason is without signal of uttering long and high-pitched sounds.
2, receiving end: answering end after the voice signal that network gets that transmitting terminal is sent, and is decoded as PCM letter to signal
Number, and signal is done into processing identical with transmitting terminal；Treated by general without signal write-in loudspeaker apparatus broadcasting of uttering long and high-pitched sounds, can be defeated
Out without sound of uttering long and high-pitched sounds.
Referring to fig. 4 which shows a untreated radio voice system is uttered long and high-pitched sounds the spectrum signature of audio sample；
Fig. 5 is one and passes through spectrum signature of the radio voice system without audio sample of uttering long and high-pitched sounds；By comparison it is found that the howl that this example provides
Make sound processing scheme that there is ideal utter long and high-pitched sounds and noise treatment effect.
Meanwhile the sound processing scheme of uttering long and high-pitched sounds is in application, only addition is handled i.e. after transmitting terminal recording, before receiving end plays
Can, do not influence the hardware or software design of existed system completely.
The basic principles, main features and advantages of the present invention have been shown and described above.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (9)
1. the processing method that voice is uttered long and high-pitched sounds in audio system, which is characterized in that the processing method passes through detection voice sound and howl
Cry sound in the feature of frequency domain, it is determined whether sound of uttering long and high-pitched sounds, and its all frequency domain components is removed when sound of uttering long and high-pitched sounds occurs；Include:
Voice signal is handled in real time in frequency domain, analysis checks voice signal whether believe by the normal voice containing voice sound
Number, it whether to be uniformly distributed in voice frequency domain by detecting voice signal, is uniformly then the normal speech signals containing voice sound,
It otherwise is background noise or sound of uttering long and high-pitched sounds；When not detecting human voice signal, it is determined as background noise or signal of uttering long and high-pitched sounds, in frequency domain
Time-domain transmission is converted back after removing the signal.
2. the processing method that voice is uttered long and high-pitched sounds in audio system according to claim 1, which is characterized in that the processing method
By checking strength characteristic of the voice signal in voice frequency domain, determine whether uniformly signal is distributed in voice frequency domain.
3. the processing method that voice is uttered long and high-pitched sounds in audio system according to claim 2, which is characterized in that the processing method
For voice signal in voice frequency domain every band frequency component, the energy intensity of each band frequency component is traversed, when energy is powerful
The counting when intensity measured in advance determines that voice signal is normal when counting, which reaches the effective voice measured in advance, to be counted
Voice signal, it is without any processing；When counting, which does not reach the effective voice measured in advance, to be counted, illustrate point of voice signal
Cloth is uneven, is determined as signal of uttering long and high-pitched sounds, and resets to frequency-region signal and handles and be converted back into time domain to eliminate signal of uttering long and high-pitched sounds.
4. the processing method that voice is uttered long and high-pitched sounds in audio system according to claim 1 or 2 or 3, which is characterized in that the place
The position of voice frequency domain is determined in reason method by following formula:
Low=(int) (size*1000f*2/rate),
Wherein, low is the position of people's acoustic low-pass, and rate is input audio specimen sample rate, and size is sample size.
5. the processing system that voice is uttered long and high-pitched sounds in audio system, which is characterized in that the processing system includes:
FFT module, the FFT module are used to audio sample progress Fast Fourier Transform going to frequency domain；
Sound detection module, the sound detection module carry out sound detection to the signal for transforming to frequency domain, determine signal in people
Whether distribution is uniform in Audio Frequency Domain；
Sound processing module, the sound processing module eliminate to signal and be uttered long and high-pitched sounds according to the testing result of sound detection module
Processing；
IFFT module, the IFFT module carry out inverse-Fourier transform to the signal handled through sound processing module and convert signal
Return to time domain.
6. the processing system that voice is uttered long and high-pitched sounds in audio system according to claim 5, which is characterized in that the sound detection
Module determines whether signal is distributed in voice frequency domain by checking strength characteristic of the voice signal in voice frequency domain
It is even.
7. the processing system that voice is uttered long and high-pitched sounds in audio system according to claim 6, which is characterized in that the sound detection
Module check voice signal every band frequency component in voice frequency domain, and the energy intensity of every band frequency component is measured with preparatory
Intensity contrast, be higher than preparatory measurement intensity when count；When counting, which reaches the effective voice measured in advance, to be counted, letter is determined
It number is evenly distributed in voice frequency domain；When counting the effective voice not up to measured in advance counting, determine signal in voice frequency
It is distributed and concentrates in domain.
8. the processing system that voice is uttered long and high-pitched sounds in audio system according to claim 5, which is characterized in that the acoustic processing
Module is not processed the signal being evenly distributed in voice frequency domain；To signal unevenly distributed in voice frequency domain, remove
Its all frequency domain components.
9. the processing system that voice is uttered long and high-pitched sounds in audio system according to claim 5, which is characterized in that the processing system
In further include initialization module, the initialization module is to determine the position of voice frequency domain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610946477.8A CN106453762B (en) | 2016-11-02 | 2016-11-02 | The processing method and system that voice is uttered long and high-pitched sounds in audio system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610946477.8A CN106453762B (en) | 2016-11-02 | 2016-11-02 | The processing method and system that voice is uttered long and high-pitched sounds in audio system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106453762A CN106453762A (en) | 2017-02-22 |
| CN106453762B true CN106453762B (en) | 2019-05-07 |
Family
ID=58178427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610946477.8A Active CN106453762B (en) | 2016-11-02 | 2016-11-02 | The processing method and system that voice is uttered long and high-pitched sounds in audio system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106453762B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108053833A (en) * | 2017-11-30 | 2018-05-18 | 广东欧珀移动通信有限公司 | Processing method, device, electronic equipment and the storage medium that voice is uttered long and high-pitched sounds |
| CN108449498B (en) * | 2018-03-12 | 2020-04-17 | Oppo广东移动通信有限公司 | Voice call data processing method and device, storage medium and mobile terminal |
| CN108449492B (en) * | 2018-03-12 | 2020-05-12 | Oppo广东移动通信有限公司 | Voice call data processing method and device, storage medium and mobile terminal |
| CN111294446B (en) * | 2019-03-28 | 2021-05-18 | 紫光展讯通信(惠州)有限公司 | Power supply system circuit of terminal |
| CN113316075B (en) * | 2021-05-11 | 2023-02-14 | 紫光展锐(重庆)科技有限公司 | Howling detection method and device and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1167391A (en) * | 1996-06-03 | 1997-12-10 | 三菱电机株式会社 | Noise-reducing talking device and method |
| KR20060069689A (en) * | 2004-12-18 | 2006-06-22 | 주식회사 팬택앤큐리텔 | Apparatus for eliminating noise of the mobile communication terminal |
| CN102316395A (en) * | 2010-07-09 | 2012-01-11 | 廖明忠 | Method and device for judging and eliminating howlround |
| CN102859591A (en) * | 2010-04-12 | 2013-01-02 | 瑞典爱立信有限公司 | Method and arrangement for noise cancellation in a speech encoder |
| CN103426433A (en) * | 2012-05-14 | 2013-12-04 | 宏达国际电子股份有限公司 | Noise cancellation method |
| CN105812993A (en) * | 2014-12-29 | 2016-07-27 | 联芯科技有限公司 | Howling detection and suppression method and device |
-
2016
- 2016-11-02 CN CN201610946477.8A patent/CN106453762B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1167391A (en) * | 1996-06-03 | 1997-12-10 | 三菱电机株式会社 | Noise-reducing talking device and method |
| KR20060069689A (en) * | 2004-12-18 | 2006-06-22 | 주식회사 팬택앤큐리텔 | Apparatus for eliminating noise of the mobile communication terminal |
| CN102859591A (en) * | 2010-04-12 | 2013-01-02 | 瑞典爱立信有限公司 | Method and arrangement for noise cancellation in a speech encoder |
| CN102316395A (en) * | 2010-07-09 | 2012-01-11 | 廖明忠 | Method and device for judging and eliminating howlround |
| CN103426433A (en) * | 2012-05-14 | 2013-12-04 | 宏达国际电子股份有限公司 | Noise cancellation method |
| CN105812993A (en) * | 2014-12-29 | 2016-07-27 | 联芯科技有限公司 | Howling detection and suppression method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106453762A (en) | 2017-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106453762B (en) | The processing method and system that voice is uttered long and high-pitched sounds in audio system | |
| US8284947B2 (en) | Reverberation estimation and suppression system | |
| CN104157295B (en) | For detection and the method for transient suppression noise | |
| US8977545B2 (en) | System and method for multi-channel noise suppression | |
| CN109637552A (en) | A kind of method of speech processing for inhibiting audio frequency apparatus to utter long and high-pitched sounds | |
| CN101583996B (en) | A method and noise suppression circuit incorporating a plurality of noise suppression techniques | |
| CN104485114B (en) | A kind of method of the voice quality objective evaluation based on auditory perception property | |
| CN101430882B (en) | Method and apparatus for restraining wind noise | |
| CN105657110B (en) | Echo cancellation method and device for voice communication | |
| CA2766196C (en) | Apparatus, method and computer program for controlling an acoustic signal | |
| CN109493877B (en) | Voice enhancement method and device of hearing aid device | |
| WO2015078121A1 (en) | Audio signal quality detection method and device | |
| CN103440869A (en) | Audio-reverberation inhibiting device and inhibiting method thereof | |
| US8306821B2 (en) | Sub-band periodic signal enhancement system | |
| CN109461455A (en) | It is a kind of to eliminate the system and method uttered long and high-pitched sounds | |
| US11562724B2 (en) | Wind noise mitigation systems and methods | |
| WO2015058484A1 (en) | Noise reducing earphone and noise reducing method thereof | |
| CN103280225B (en) | Low-complexity silence detection method | |
| CN107180643A (en) | One kind is uttered long and high-pitched sounds sound detection and elimination system | |
| CN101808260A (en) | Audio dynamic feedback suppression method | |
| CN111182431A (en) | Howling suppression method for conference sound reinforcement system | |
| CN103796136A (en) | Equipment and method for ensuring output loudness and tone quality of different sound effect modes | |
| US11153684B2 (en) | Dynamic debuzzer for speakers | |
| CN105872884A (en) | Headset as well as downlink noise reduction circuit and method for headset | |
| Rahman et al. | Low-frequency band noise suppression using bone conducted speech |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Processing method and system of voice scream in audio system Effective date of registration: 20211217 Granted publication date: 20190507 Pledgee: Societe Generale Bank Co.,Ltd. Qingpu Branch of Shanghai Pledgor: SHANGHAI SHUGUO TECHNOLOGY Co.,Ltd. Registration number: Y2021310000141 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20221207 Granted publication date: 20190507 Pledgee: Societe Generale Bank Co.,Ltd. Qingpu Branch of Shanghai Pledgor: SHANGHAI SHUGUO TECHNOLOGY Co.,Ltd. Registration number: Y2021310000141 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Processing Method and System of Voice Scream in Audio System Effective date of registration: 20221222 Granted publication date: 20190507 Pledgee: Societe Generale Bank Co.,Ltd. Qingpu Branch of Shanghai Pledgor: SHANGHAI SHUGUO TECHNOLOGY Co.,Ltd. Registration number: Y2022980028673 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20231225 Granted publication date: 20190507 Pledgee: Societe Generale Bank Co.,Ltd. Qingpu Branch of Shanghai Pledgor: SHANGHAI SHUGUO TECHNOLOGY Co.,Ltd. Registration number: Y2022980028673 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: The processing method and system of voice howling in audio systems Effective date of registration: 20231227 Granted publication date: 20190507 Pledgee: Societe Generale Bank Co.,Ltd. Qingpu Branch of Shanghai Pledgor: SHANGHAI SHUGUO TECHNOLOGY Co.,Ltd. Registration number: Y2023980074572 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |