CN1956480A - Acoustic echo removing method - Google Patents
Acoustic echo removing method Download PDFInfo
- Publication number
- CN1956480A CN1956480A CN 200510030841 CN200510030841A CN1956480A CN 1956480 A CN1956480 A CN 1956480A CN 200510030841 CN200510030841 CN 200510030841 CN 200510030841 A CN200510030841 A CN 200510030841A CN 1956480 A CN1956480 A CN 1956480A
- Authority
- CN
- China
- Prior art keywords
- echo
- acoustic echo
- filtering
- removing method
- acoustic
- 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
Images
Landscapes
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
Abstract
This invention relates to a method for eliminating acoustic echo including the following steps: the minimum mean square algorithm of adaptive step wave band standardization, eliminating wave beams of noises to form non-linear filter to eliminate acoustic echoes. This invention applies an exclusive adaptive step non-linear filter technology to divide signals into multiple wave bands for process and alter steps adaptively according to the echo environment to eliminate echoes to increase convergence speed and reduce radiation speed to improve the ability for eliminating echo.
Description
Technical field
The present invention relates to a kind of echo cancel method, especially refer to eliminate the method that acoustic echo is eliminated in the echo technology.
Background technology
Acoustic echo is meant after sound that loudspeaker plays is come out is picked up by microphone beams back far-end, and this just makes far end talk person can hear the sound of oneself.Acoustic echo is divided into direct echo and indirect echo again.Directly echo is meant that the sound that loudspeaker plays is come out directly enters microphone without any reflection.This echo delay is the shortest, and the broadcast sound volume of the distance between the speech energy of it and remote speaker, loud speaker and microphone, angle, loud speaker and the factors such as pick-up sensitivity of microphone are relevant.Echo is meant that the sound of loudspeaker plays enters the echo set that microphone produces after different path one or many reflections indirectly.Because the change of object on every side, walking about etc. of people for example all can change the return path of echo, because the characteristics of this echo are to become in multipath, time.In addition, background noise also is one of echogenic factor.
See also Fig. 1, this is a multi-media voice environment schematic diagram.The important method that acoustic echo is eliminated is to use acoustic echo canceller (AEC:Acoustic Echo Cancellation), AEC is based on the correlation of loudspeaker signal with the multipath echo that is produced by it, set up the speech model of remote signaling, utilize it that echo is estimated, and constantly revise the coefficient of filter, make estimated value approach real echo more.Then, the echo estimated value is deducted from the input signal of microphone, thereby reach the purpose of eliminating echo, AEC also compares the input of microphone with the value of loud speaker output, prolongs the repeatedly acoustic echo of reflection that postpones thereby eliminate.What of the output valve of the loud speaker in the past that root a tree name memory is deposited, AEC can eliminate the echo of various delays.
General echo cancellation mainly contains at present: the one, and the application scheme of hands-free phone is promptly only done gain adjustment, does not have the elimination of real echo.When a side talked, the gain raising with its signal transmission direction reduced its rightabout gain simultaneously, to reduce echo.Consequently when both sides talk simultaneously, have only a side sound to send out, have only one can hear the other side, or both sides' choppy voice.Another kind is to adopt the echo of lowest mean square (LMS) to eliminate.Traditional lowest mean square convergence is slow, easily disperses, and can only eliminate remaining echo, have to eliminate remaining echo with nonlinear method, though this method is more effective than the gain adjusting method, still have many weak points, the happening occasionally intermittently of echo or sound.
Summary of the invention
The acoustic echo removing method that the object of the present invention is to provide a kind of high-performance and be easy to realize, thus reach the effect of noise reduction and high-fidelity voice.
A kind of acoustic echo removing method provided by the present invention comprises the following steps: adaptive step subrane standardization least mean square algorithm; The wave beam of eliminating noise forms nonlinear filtering; The realization acoustic echo is eliminated.
In above-mentioned acoustic echo removing method, the wave beam of eliminating noise forms nonlinear filtering and comprises following process: the microphone signal carries out mould/number conversion → filtering → time domain/frequency domain conversion → big echo elimination → wave beam and forms echo elimination → post-filtering → smothing filtering → frequency domain/time domain conversion → filtering → D/A switch → power amplifier → loud speaker.
In above-mentioned acoustic echo removing method, adaptive step subrane standardization least mean square algorithm is to change step-length, the also adaptive variation of the speed of change according to echo environment self-adaption ground.
In above-mentioned acoustic echo removing method, step-length has only when near-end is talked, and it is minimum; When both sides talked simultaneously, it was adjusted into moderate-length; When having only the far-end speech, it is a maximum.
In above-mentioned acoustic echo removing method, the wave beam of eliminating noise forms nonlinear filtering by signal is carried out analyzing and testing, identifies background noise, filters background noise with the dynamic adaptive filter technology then.
Owing to adopted above-mentioned technical solution, promptly adopt unique adaptive step nonlinear filtering wave technology, become a plurality of wave bands to handle respectively signal segmentation, change step-length according to echo environment self-adaption ground, eliminate filtering through echo, can improve convergence rate like this, reduce divergence speed, can improve the ability that echo is eliminated simultaneously.The present invention had both solved echo problem, the speech quality of Bao Zhenging simultaneously, and have good full-duplex preformance.The present invention is applicable to fields such as automobile hands-free phone, notebook computer, PDA, USB portable handsfree phone, conference telephone, video telephone, mobile phone, cordless telephone, portable DVD player.
Description of drawings
Fig. 1 is existing multi-media voice environment schematic diagram;
Fig. 2 is being not intended to of acoustic echo removing method of the present invention;
Fig. 3 is the schematic diagram of one of acoustic echo removing method embodiment of the present invention.
Embodiment
Referring to Fig. 2, this is the schematic diagram that adopts acoustic echo removing method of the present invention.
The present invention is used for acoustic echo and eliminates, and adopts unique adaptive step nonlinear filtering wave technology, promptly adopts at adaptive step subrane standardization least mean square algorithm, and adopts the wave beam of eliminating noise to form the nonlinear filtering wave technology.
Suppose: h (n) is the echo channel response, and X (n) is a reference signal, and C (n) is the adaptive-filtering coefficient, and E (n) is the signal of microphone receiving terminal.
E (n)=X (n) * h (n) (convolution)
E (n)=EL (n) (linearity)+EN (n) (non-linear)
C(n+1)＝C(n)+ΔC(n)
Δ C (n)=μ X (n) e (n)/‖ x (n) ‖
2(μ is an adaptive step)
(n)=X (n) * C (n) (convolution)
e(n)＝E(n)-(n)
In order to eliminate echo, acoustic echo is eliminated the function must possess the acoustic response of dynamic learning loud speaker under the acoustic enviroment between the microphone, become a plurality of wave bands to handle respectively signal segmentation, change step-length according to echo environment self-adaption ground, eliminate filtering technique by echo, can improve convergence rate like this, reduce divergence speed, can improve the ability that echo is eliminated simultaneously.
Referring to Fig. 3, the present invention is used for acoustic echo and eliminates, adopt adaptive step subrane standardization least mean square algorithm, its wave beam of eliminating noise forms nonlinear filtering and mainly comprises: big Echo Canceller, wave beam form Echo Canceller, postfilter, smoothing filter, handle other front-end and back-end, comprises continuously with discrete signal conversion, time-domain and frequency-domain conversion of signals, uses convectional signals processing method such as filtering always.
The present invention is work like this: adopting unique adaptive step nonlinear filtering wave technology is not to train Echo Canceller with noise, but the sound that exchanges according to normal talk continuous improve Echo Canceller, it is more natural to make meeting begin the speech of speaking unceasingly the back.
The echo that good acoustic echo canceller is handled is on close level in the echo level of reality and should be suddenlyd change fast and not, and this depends on the degree fast when it is in accepting state and adapts in this state.Adaptive step subrane standardization least mean square algorithm changes step-length, the also adaptive variation of the speed of change according to echo environment self-adaption ground.When only when near-end is talked, step-length is minimum; When both sides talk simultaneously (during full duplex), step-length is adjusted into moderate-length; When having only the far-end speech, step-length is a maximum.This scheme is switched system fast in different states, and keeps stable, still has fabulous performance when echo path is changed.The wave beam of eliminating noise forms nonlinear filter by signal is carried out analyzing and testing, identifies background noise, filters background noise with the dynamic adaptive filter technology then, but does not damage voice signal, thereby reach the effect of noise reduction and high-fidelity voice; Evaluation is adjusted the gain of voice signal automatically by voice and background noise are carried out analyzing and testing, makes its output remain on certain level.
The concrete grammar (as Fig. 3 content) that adopts echo of the present invention to eliminate, its sidetone is eliminated can reach 30dB～40dB, can be controlled at simultaneously under the pattern of both sides' speech to carry out level and smooth transition, makes voice true, undistorted, and adaptation fast and accurately.
In sum, the present invention is used for acoustic echo and eliminates, and acoustic echo offsets can be up to dB more than 60 (being international best level, than the high about 20dB of like product) at present.Use other microphone (being used for reference signal picks up), can obtain more superior full-duplex preformance and echo cancellation performance.The present invention realizes acoustic echo elimination function, and can reach the effect of noise reduction and high-fidelity voice, is applied in the Echo Canceller scheme, and is very practical.
Claims (5)
1. an acoustic echo removing method comprises the following steps:
Adaptive step subrane standardization least mean square algorithm;
The wave beam of eliminating noise forms nonlinear filtering;
The realization acoustic echo is eliminated.
2. acoustic echo removing method as claimed in claim 1 is characterized in that: the wave beam of described elimination noise forms nonlinear filtering and comprises following process:
The microphone signal carries out mould/number conversion → filtering → time domain/frequency domain conversion → big echo elimination → wave beam and forms echo elimination → post-filtering → smothing filtering → frequency domain/time domain conversion → filtering → D/A switch → power amplifier → loud speaker.
3. acoustic echo removing method as claimed in claim 1 is characterized in that: described adaptive step subrane standardization least mean square algorithm is to change step-length, the also adaptive variation of the speed of change according to echo environment self-adaption ground.
4. acoustic echo removing method as claimed in claim 3 is characterized in that: described step-length has only when near-end is talked, and it is minimum; When both sides talked simultaneously, it was adjusted into moderate-length; When having only the far-end speech, it is a maximum.
5. acoustic echo removing method as claimed in claim 1 is characterized in that: the wave beam of described elimination noise forms nonlinear filtering by signal is carried out analyzing and testing, identifies background noise, filters background noise with the dynamic adaptive filter technology then.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510030841 CN1956480A (en) | 2005-10-28 | 2005-10-28 | Acoustic echo removing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510030841 CN1956480A (en) | 2005-10-28 | 2005-10-28 | Acoustic echo removing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1956480A true CN1956480A (en) | 2007-05-02 |
Family
ID=38063560
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200510030841 Pending CN1956480A (en) | 2005-10-28 | 2005-10-28 | Acoustic echo removing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1956480A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101888455A (en) * | 2010-04-09 | 2010-11-17 | 熔点网讯(北京)科技有限公司 | Self-adaptive echo counteracting method for frequency domain |
| CN102413384A (en) * | 2011-11-16 | 2012-04-11 | 杭州艾力特音频技术有限公司 | Echo cancellation two-way voice talk back equipment |
| CN102655558A (en) * | 2012-05-21 | 2012-09-05 | 宁波工程学院 | Double-end pronouncing robust structure and acoustic echo cancellation method |
| CN104835502A (en) * | 2015-05-20 | 2015-08-12 | 捷思锐科技（北京）有限公司 | Audio signal processing method, audio signal processing device and electronic equipment |
| CN105391879A (en) * | 2015-12-09 | 2016-03-09 | 天津大学 | Echo residue-free double-end communication robust acoustic echo elimination method |
| CN105491256A (en) * | 2015-12-09 | 2016-04-13 | 天津大学 | Robust step size adjustment method in initialization phase of acoustic echo cancellation |
| US9832582B2 (en) | 2013-09-16 | 2017-11-28 | Huawei Device (Dongguan) Co., Ltd. | Sound effect control method and apparatus |
| CN107636758A (en) * | 2015-05-15 | 2018-01-26 | 哈曼国际工业有限公司 | Acoustic echo eliminates system and method |
| CN110211602A (en) * | 2019-05-17 | 2019-09-06 | 北京华控创为南京信息技术有限公司 | Intelligent sound enhances communication means and device |
-
2005
- 2005-10-28 CN CN 200510030841 patent/CN1956480A/en active Pending
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101888455A (en) * | 2010-04-09 | 2010-11-17 | 熔点网讯(北京)科技有限公司 | Self-adaptive echo counteracting method for frequency domain |
| CN101888455B (en) * | 2010-04-09 | 2013-07-03 | 熔点网讯(北京)科技有限公司 | Self-adaptive echo counteracting method for frequency domain |
| CN102413384A (en) * | 2011-11-16 | 2012-04-11 | 杭州艾力特音频技术有限公司 | Echo cancellation two-way voice talk back equipment |
| CN102655558A (en) * | 2012-05-21 | 2012-09-05 | 宁波工程学院 | Double-end pronouncing robust structure and acoustic echo cancellation method |
| CN102655558B (en) * | 2012-05-21 | 2013-10-09 | 宁波工程学院 | Double-end pronouncing robust structure and acoustic echo cancellation method |
| US9832582B2 (en) | 2013-09-16 | 2017-11-28 | Huawei Device (Dongguan) Co., Ltd. | Sound effect control method and apparatus |
| CN107636758A (en) * | 2015-05-15 | 2018-01-26 | 哈曼国际工业有限公司 | Acoustic echo eliminates system and method |
| CN107636758B (en) * | 2015-05-15 | 2022-05-24 | 哈曼国际工业有限公司 | Acoustic echo cancellation system and method |
| CN104835502A (en) * | 2015-05-20 | 2015-08-12 | 捷思锐科技（北京）有限公司 | Audio signal processing method, audio signal processing device and electronic equipment |
| CN104835502B (en) * | 2015-05-20 | 2018-04-10 | 北京捷思锐科技股份有限公司 | Acoustic signal processing method, device and electronic equipment |
| CN105391879A (en) * | 2015-12-09 | 2016-03-09 | 天津大学 | Echo residue-free double-end communication robust acoustic echo elimination method |
| CN105491256A (en) * | 2015-12-09 | 2016-04-13 | 天津大学 | Robust step size adjustment method in initialization phase of acoustic echo cancellation |
| CN105391879B (en) * | 2015-12-09 | 2018-06-12 | 天津大学 | A kind of acoustic echo removing method of echoless residual dual end communication robust |
| CN105491256B (en) * | 2015-12-09 | 2018-12-04 | 天津大学 | A kind of acoustic echo canceller startup stage steady step length regulating method |
| CN110211602A (en) * | 2019-05-17 | 2019-09-06 | 北京华控创为南京信息技术有限公司 | Intelligent sound enhances communication means and device |
| CN110211602B (en) * | 2019-05-17 | 2021-09-03 | 北京华控创为南京信息技术有限公司 | Intelligent voice enhanced communication method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1956480A (en) | Acoustic echo removing method | |
| CN108376548B (en) | Echo cancellation method and system based on microphone array | |
| US11297178B2 (en) | Method, apparatus, and computer-readable media utilizing residual echo estimate information to derive secondary echo reduction parameters | |
| US6961422B2 (en) | Gain control method for acoustic echo cancellation and suppression | |
| US9451078B2 (en) | Universal reconfigurable echo cancellation system | |
| CN101562669B (en) | Method of adaptive full duplex full frequency band echo cancellation | |
| US9992572B2 (en) | Dereverberation system for use in a signal processing apparatus | |
| CN1223166C (en) | Methods and apparatus for improved sub-band adaptive filtering in echo cancellation systems | |
| US10115412B2 (en) | Signal processor with side-tone noise reduction for a headset | |
| US20070019803A1 (en) | Loudspeaker-microphone system with echo cancellation system and method for echo cancellation | |
| CN1533665A (en) | Echo canceller having nonlinear echo suppressor for harmonics calculations | |
| WO2012142270A1 (en) | Systems, methods, apparatus, and computer readable media for equalization | |
| CN104052526A (en) | Clipping Based On Cepstral Distance For Acoustic Echo Canceller | |
| EP1727344A3 (en) | Improved echo cancellation in telephones with multiple microphones | |
| CN1890892A (en) | Echo canceller having a series arrangement of adaptive filters with individual update control strategy | |
| CN107483761A (en) | A kind of echo suppressing method and device | |
| US9020144B1 (en) | Cross-domain processing for noise and echo suppression | |
| EP2101480A3 (en) | Echo canceller and echo cancelling method | |
| CN109273019A (en) | Method and echo inhibition for the dual call detection that echo inhibits | |
| US10938992B1 (en) | Advanced audio feedback reduction utilizing adaptive filters and nonlinear processing | |
| CN1209889C (en) | Phoneme background noise inhibitor | |
| US20090067615A1 (en) | Echo cancellation using gain control | |
| CN111917926A (en) | Echo cancellation method and device in communication terminal and terminal equipment | |
| CN115579016A (en) | Method and system for eliminating acoustic echo | |
| CN1663230A (en) | Non stationary echo canceller |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20070502 |