CN1219391C - Echo eliminator and method for evaluating echo - Google Patents
Echo eliminator and method for evaluating echo Download PDFInfo
- Publication number
- CN1219391C CN1219391C CN 02146714 CN02146714A CN1219391C CN 1219391 C CN1219391 C CN 1219391C CN 02146714 CN02146714 CN 02146714 CN 02146714 A CN02146714 A CN 02146714A CN 1219391 C CN1219391 C CN 1219391C
- Authority
- CN
- China
- Prior art keywords
- echo
- input signal
- signal
- end input
- remote
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
Abstract
The present invention discloses an echo eliminator which comprises an adaptive filter, a remote-end input signal buffer memory, a near-end input signal buffer memory and a signal adder. In the present invention, a drop sampling processing module for remote-end signals, a drop sampling processing module for near-end signals and a searching module for a cross correlation echo effective part are added, wherein the drop sampling processing module for remote-end signals and the drop and sampling processing module for near-end signals respectively carry out drop sampling to remote-end input signals and near-end input signals to obtain echo effective parameters; the signal adder carries out echo counteraction by input signal mixed with echo signals received from the external of the echo eliminator subtracting an echo estimation value produced by the adaptive filter. The present invention discloses an echo estimation method simultaneously, which comprises the following steps: firstly, the drop sampling is respectively carried out to the remote-end input signals and the near-end input signals; then, the effective parameters of echo are solved; an echo value is calculated through an adaptive algorithm. The proposal of the present invention applied has the advantages that the counteraction effect of echo is not influenced; the efficiency is enhanced; the system cost is saved by large amount.
Description
Technical field
The present invention relates in the communication system to the treatment technology of echo particularly a kind of Echo Canceller and echo method of estimation.
Background technology
In the Fixed Telephony system, switch is inner, and what use is the system of four lines, is two-wire system and user's telephone wire uses, like this, needs to realize the transmission of signal from the switch to user by a transducer (Hybrid) in transmission course.Because the impedance of four-wire system and two-wire system can not be accomplished accurate match, signal reflection will occur in the process of transmission, and the signal of reflection arrives the user side user and will hear that own sound has in a minute returned again and come, and has formed echo.Echo will be reduced speech quality greatly, be a problem that must solve in communication system.Signal as A end among Fig. 1 turns back to the A end again through transducer 101, thereby the party A-subscriber has heard the echo of oneself.
At present, can realize eliminating the function of echo by the method for using Echo Canceller.Fig. 2 is the structured flowchart of prior art Echo Canceller, and as shown in Figure 2, the prior art Echo Canceller comprises a sef-adapting filter 202, remote end input signal buffer memory 201, a near end input signal buffer memory 203 and a signal adder 204.Wherein far-end refers to four line ends, and near-end is meant two line ends.Its core is a sef-adapting filter 202, by come the transmission characteristic of analog converter with sef-adapting filter 202, near end input signal that this sef-adapting filter 202 sends with reference to the remote end input signal that is sent by remote end input signal buffer memory 201 and near end input signal buffer memory 203 and the value of coming estimated echo as the near-end output signal of the feedback signal of sef-adapting filter, the effect of sef-adapting filter 202 is exactly to imitate out an environment that is similar to real reflection path (echopath), and it makes the echo that estimates near true echo in can how long itself, the difference of this echo that estimates and true echo is less than a setting, to deduct the echo value of estimation from the outside signal that is mixed with echo that receives of Echo Canceller by signal adder 204 then, carry out echo cancelltion.The difference of echo of estimating and true echo is more little, and the effect of echo cancelltion is just good more.
Fig. 3-1 and Fig. 3-2 is echogenic principle schematic.Wherein, Fig. 3-1 is for sending the schematic diagram of ideal pulse pumping signal, the response schematic diagram of Fig. 3-2 ideal pulse pumping signals that sent for Fig. 3-1, for the point of the A among Fig. 3-1, a desirable impact signal that sends at the A point will obtain the response among Fig. 3-2, wherein, tp is clean echo segment, be the echo live part, tr is pure time-delay part, and td is total echo time-delay.Shown in Fig. 3-2, clean echo segment is the less part of accounting in total echo time-delay.As seen,, just can realize the work that echo is eliminated, yet present Echo Canceller all is that omnidistance time td segment signal is offset if clean echo segment effectively can be offset.Obviously, the result who does has like this brought system resource waste in large quantities.
As seen, when the major defect of prior art carries out echo cancelltion exactly, do not consider the characteristic of echo, directly the signal of omnidistance time period of echo is offset,, and cause to improve resource utilization ratio and reduce cost so the resource that needs is too many.
Summary of the invention
The basic principle that is produced by echo shows that the live part of echo only accounts for very little one in whole echo length, if can find this part, just can use less resource to realize the effect of echo cancelltion, thereby can save system cost in large quantities.
In view of this, main purpose of the present invention is to provide a kind of Echo Canceller, can find the live part of echo, and only this part echo is offset, not only do not influence the effect of echo cancelltion, and improved efficient, can save system cost in large quantities.
Another main purpose of the present invention is to provide a kind of echo method of estimation, can obtain the echo estimated value efficiently and accurately.
An aspect according to above-mentioned purpose the invention provides a kind of Echo Canceller, and it comprises sef-adapting filter, remote end input signal buffer memory, near end input signal buffer memory and signal adder; This remote end input signal buffer memory sends to sef-adapting filter with the remote end input signal that receives as the reference signal; This near end input signal buffer memory sends to sef-adapting filter with the near end input signal that receives; This sef-adapting filter echogenicity estimated value also sends it to signal adder; The input signal that is mixed with echo signal that this signal adder will receive from the Echo Canceller outside deducts the echo estimated value and carries out echo cancelltion, and produces the near-end output signal, and this near-end output signal is sent to sef-adapting filter as feedback signal; This Echo Canceller further comprises:
The down-sampled processing module of remote signaling, it will carry out down-sampled processing from the remote end input signal that the remote end input signal buffer memory receives, and the signal after will handling sends to cross-correlation echo live part search module;
The down-sampled processing module of near end signal, it will carry out down-sampled processing from the near end input signal that the near end input signal buffer memory receives, and the signal after will handling sends to cross-correlation echo live part search module;
Cross-correlation echo live part search module, signal after above-mentioned two kinds of processing that it will be received carries out computing cross-correlation, from result of calculation, find out the echo maximum, in the echo maximum left and right sides according to communication protocol, get distance part farthest as the echo live part, it is sent to sef-adapting filter as effective echo parameter;
Sef-adapting filter receives the signal and the near-end output signal of remote end input signal buffer memory and the transmission of near end input signal buffer memory, and receive effective echo parameter, find effective echo segment in the received signal according to effective echo parameter, effective echo segment to the above-mentioned signal that receives carries out echo estimation, echogenicity estimated value.
Wherein, down-sampled processing module of remote signaling or the down-sampled processing module of near end signal can further comprise remote signaling low pass filter or near end signal low pass filter, be respectively applied for and carry out low-pass filtering from the remote end input signal or the near end input signal of remote end input signal buffer memory or the reception of near end input signal buffer memory, carry out down-sampled processing again.
Remote signaling low pass filter and near end signal low pass filter can be respectively 1000 hertz low pass filter for cut-off frequency.
According to another aspect of above-mentioned purpose, the invention provides a kind of echo method of estimation, adopt above-mentioned Echo Canceller to carry out echo estimation, this method may further comprise the steps:
1) down-sampled processing module of remote signaling and the down-sampled processing module of near end signal are carried out down-sampled to remote end input signal and near end input signal respectively;
2) will send to cross-correlation echo live part search module through two kinds of down-sampled signals, carry out computing cross-correlation;
3) cross-correlation echo live part search module is found out the echo maximum from cross correlation results, in the echo maximum left and right sides according to communication protocol, get distance part farthest as the echo live part, it as the echo actual parameter, is sent to sef-adapting filter;
4) sef-adapting filter finds effective echo segment in the received signal according to effective echo parameter, and effective echo segment of the remote end input signal, near end input signal and the near-end output signal that receive is carried out echo estimation, echogenicity estimated value.Wherein, this method step 1) may further include: remote end input signal and near end input signal are carried out the down-sampled low-pass filtering of carrying out respectively earlier before.Step 3) is described find out the peaked method of echo can for: compare by each some amplitude and to search out the echo maximum cross correlation results.
By such scheme as can be seen, key of the present invention is: carry out down-sampled respectively to remote end input signal and near end input signal earlier; Again above-mentioned signal is carried out computing cross-correlation; From cross correlation results, search out the echo live part, obtain the echo actual parameter; Sef-adapting filter parameter Estimation in view of the above goes out the echo value.Therefore, this Echo Canceller of the present invention and echo method of estimation can be located effective echo segment exactly, only the live part to this echo carries out echo cancelltion, not only do not influence the effect of echo cancelltion, and can raise the efficiency, and save system cost in a large number.Because the present invention has adopted down-sampled method, so in the search of delaying time, can reduce the memory space of amount of calculation and data.
Description of drawings
Fig. 1 is an echogenic system schematic in the prior art communication system;
Fig. 2 is the structured flowchart of prior art Echo Canceller;
Fig. 3-1 is for sending the schematic diagram of ideal pulse pumping signal;
The response schematic diagram of Fig. 3-2 ideal pulse pumping signals that sent for Fig. 3-1;
Fig. 4 is the structured flowchart of a preferred embodiment of Echo Canceller of the present invention;
Fig. 5 is the applications schematic diagram of Echo Canceller of the present invention;
Fig. 6 carries out the principle schematic of computing cross-correlation for the present invention;
Fig. 7 carries out the process schematic diagram that echo is estimated for the present invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, by the following examples, and with reference to accompanying drawing, the present invention is described in more detail.
Fig. 4 is the structured flowchart of a preferred embodiment of Echo Canceller of the present invention, as shown in Figure 4, Echo Canceller of the present invention comprises: the down-sampled processing module of the down-sampled processing module of sef-adapting filter 402, remote end input signal buffer memory 401, near end input signal buffer memory 406, remote signaling 403, near end signal 405, a cross-correlation echo live part search module 404 and a signal adder 407.
Wherein, remote end input signal buffer memory 401 sends to sef-adapting filter 402 with the remote end input signal that receives as the reference signal.Near end input signal buffer memory 401 sends to sef-adapting filter 402 with the near end input signal that receives.The down-sampled processing module 403 of remote signaling will be carried out down-sampled processing from the remote end input signal that remote end input signal buffer memory 401 receives, the down-sampled processing module 403 of remote signaling can also contain a remote signaling low pass filter, it is 1000 hertz low pass filter that present embodiment has adopted a cut-off frequency, this module will be carried out low-pass filtering from the remote end input signal that remote end input signal buffer memory 401 receives earlier, carry out down-sampled processing again, and the signal after will handling sends to cross-correlation echo live part search module 404.The down-sampled processing module 405 of near end signal will be carried out down-sampled processing from the near end input signal that near end input signal buffer memory 406 receives, the down-sampled processing module 405 of near end signal can also comprise a near end signal low pass filter, it is 1000 hertz low pass filter that present embodiment has adopted a cut-off frequency, this module will be carried out low-pass filtering from the near end input signal that near end input signal buffer memory 401 receives earlier, carry out down-sampled processing again, and the signal after will handling sends to cross-correlation echo live part search module 404.Carry out computing cross-correlation after above-mentioned two kinds of processing that cross-correlation echo live part search module 404 will be received, promptly above-mentioned two kinds of signals are carried out related operation, from the result of cross-correlation, search maximum and live part in the echo then, maximum and live part according to echo produces effective echo parameter again, and this parameter is sent to sef-adapting filter 402.Sef-adapting filter 402 receives the signal of remote end input signal buffer memory 401 and 406 transmissions of near end input signal buffer memory, and as the near-end output signal of the feedback signal 402 of sef-adapting filter, receive effective echo parameter simultaneously, and according to above-mentioned signal and effective echo parameter echogenicity estimated value and send it to signal adder 407.The input signal that is mixed with echo signal that signal adder 407 will receive from the Echo Canceller outside deducts the echo estimated value and carries out echo cancelltion.
Because the signals sampling rate is all higher relatively in communication system, for example the sample rate in public circuit switched telephone network (PSTN) is 8k or the like.If the scope that system need search for is 64ms, so, the number of samples that need carry out cross-correlation calculation and search when carrying out cross-correlation just is 512 points, and wherein, every 1ms has 8 sampling points.Amount of calculation is too big like this, is unfavorable for improving resource utilization ratio.Therefore, the present invention carries out some processing before carrying out related operation.The mode of handling is down-sampled.
Down-sampled principle is the data of sampling originally, once evenly extracts again, and be the data of 8k for example to sample rate in the public circuit switched telephone network (PSTN), per four data extract one, promptly fall 4 samplings.At this moment, for the data of carrying out computing cross-correlation, be equivalent to sample rate and be reduced to 2kHz, amount of calculation has obtained very big reduction, though the spectrum energy of voice signal compares broad, but the part that its energy is relatively concentrated is below 1000Hz, because it can keep the part of original signal frequency concentration of energy, so can not influence the result of search when carrying out effective echo search.If adopt lower sample rate, influence the frequency range that speech signal energy is comparatively concentrated possibly, this can influence the search of effective echo.
In actual applications, after down-sampled, cause sample rate to cross low and spectrum leakage and spectral aliasing that cause, to such an extent as to influence result calculated, carrying out input signal being carried out low-pass filtering in down-sampled, has only 2kHz because sample rate is actual for the sampling point of cross-correlation, so what adopt is that the low pass filter that passband is not higher than 1000Hz carries out low-pass filtering.Simultaneously, remain as much as possible, be not higher than the filter of 1000Hz so adopt passband to be higher than 800Hz owing to will guarantee the frequency band that voice signal is concentrated, for example, the low pass filter of the 1000Hz that adopts in the present embodiment.
Fig. 5 is the applications schematic diagram of Echo Canceller of the present invention, and as shown in Figure 5, dotted line is the echo signal that the A end produces when the B end sends signal among the figure, and this signal is cancelled through Echo Canceller.
Fig. 6 carries out the principle schematic of computing cross-correlation for the present invention.Wherein, a is initial pumping signal, and b is a response signal, and c is the result, as shown in Figure 6, find ts by search peaked method from c as a result, by revising, the method for correction is subsequently: in the echo maximum left and right sides according to communication protocol, get distance part farthest, as echo live part tr, so just can find effective echo segment of system, just can carry out the work that echo is eliminated to effective echo segment after finding with very little resource.
Therefore, the present invention carries out the echo estimation approach, and Fig. 7 is the process schematic diagram that carries out the echo estimation embodiment illustrated in fig. 4.It may further comprise the steps:
Step 701 and 702 at first, carry out down-sampled respectively to remote end input signal and near end input signal, present embodiment is earlier remote end input signal and near end input signal to be carried out step 7011,7021 respectively, cut-off frequency is 1000 hertz a low-pass filtering, carry out step 7012,7022 again, 4 samplings are fallen in signal.
Step 703 will be carried out computing cross-correlation through two kinds of down-sampled signals then, be the cross-correlation of calculating 128 sampling points in the present embodiment.
Then, step 704 searches out echo maximum and live part from cross correlation results, obtain the echo actual parameter, be to find out the echo maximum in the present embodiment by cross correlation results each point amplitude is compared, in the echo maximum left and right sides according to communication protocol, get distance part farthest,, it is sent to sef-adapting filter as the echo actual parameter as the echo live part.
Final step 705, sef-adapting filter is according to the echo actual parameter, find effective echo segment of the signal of its reception, simultaneously estimate the echo value according to this effective echo segment, remote end input signal, near end input signal and as the near-end output signal of the feedback signal of sef-adapting filter.
This shows, this Echo Canceller of the present invention and echo method of estimation, only the live part to echo carries out echo cancelltion, does not influence the effect of echo cancelltion, has improved efficient, has saved system cost in a large number.Because the present invention has adopted down-sampled method, so in the search of delaying time, can reduce the memory space of amount of calculation and data.Simultaneously, guaranteed that signal can be because of down-sampled former thereby produce accuracy that spectral aliasing influence time-delay is estimated to such an extent as to carry out low-pass filtering before down-sampled.
Claims (8)
1, a kind of Echo Canceller, it comprises sef-adapting filter, remote end input signal buffer memory, near end input signal buffer memory and signal adder; This remote end input signal buffer memory sends to sef-adapting filter with the remote end input signal that receives as the reference signal; This near end input signal buffer memory sends to sef-adapting filter with the near end input signal that receives; This sef-adapting filter echogenicity estimated value also sends it to signal adder; The input signal that is mixed with echo signal that this signal adder will receive from the Echo Canceller outside deducts the echo estimated value and carries out echo cancelltion, and produces the near-end output signal, and this near-end output signal is sent to sef-adapting filter as feedback signal; It is characterized in that this Echo Canceller further comprises:
The down-sampled processing module of remote signaling, it will carry out down-sampled processing from the remote end input signal that the remote end input signal buffer memory receives, and the signal after will handling sends to cross-correlation echo live part search module;
The down-sampled processing module of near end signal, it will carry out down-sampled processing from the near end input signal that the near end input signal buffer memory receives, and the signal after will handling sends to cross-correlation echo live part search module;
Cross-correlation echo live part search module, signal after above-mentioned two kinds of processing that it will be received carries out computing cross-correlation, from result of calculation, find out the echo maximum, in the echo maximum left and right sides according to communication protocol, get distance part farthest as the echo live part, it is sent to sef-adapting filter as effective echo parameter;
Sef-adapting filter receives the signal and the near-end output signal of remote end input signal buffer memory and the transmission of near end input signal buffer memory, and receive effective echo parameter, find effective echo segment in the received signal according to effective echo parameter, effective echo segment to the above-mentioned signal that receives carries out echo estimation, echogenicity estimated value.
2, Echo Canceller as claimed in claim 1, it is characterized in that: the down-sampled processing module of described remote signaling further comprises the remote signaling low pass filter, after being used for to carry out low-pass filtering from the remote end input signal that the remote end input signal buffer memory receives, carry out down-sampled processing again.
3, Echo Canceller as claimed in claim 2 is characterized in that: described remote signaling low pass filter is that cut-off frequency is 1000 hertz a low pass filter.
4, Echo Canceller as claimed in claim 1, it is characterized in that: the down-sampled processing module of described near end signal further comprises the near end signal low pass filter, after being used for to carry out low-pass filtering from the near end input signal that the near end input signal buffer memory receives, carry out down-sampled processing again.
5, Echo Canceller as claimed in claim 4 is characterized in that: described near end signal low pass filter is that cut-off frequency is 1000 hertz a low pass filter.
6, a kind of echo method of estimation is characterized in that, adopts the described Echo Canceller of claim 1 to carry out echo estimation, and this method may further comprise the steps:
1) down-sampled processing module of remote signaling and the down-sampled processing module of near end signal are carried out down-sampled to remote end input signal and near end input signal respectively;
2) will send to cross-correlation echo live part search module through two kinds of down-sampled signals, carry out computing cross-correlation;
3) cross-correlation echo live part search module is found out the echo maximum from cross correlation results, in the echo maximum left and right sides according to communication protocol, get distance part farthest as the echo live part, it is sent to sef-adapting filter as the echo actual parameter;
4) sef-adapting filter finds effective echo segment in the received signal according to effective echo parameter, and effective echo segment of the remote end input signal, near end input signal and the near-end output signal that receive is carried out echo estimation, echogenicity estimated value.
7, echo estimation approach as claimed in claim 6 is characterized in that step 1) further comprises: remote end input signal and near end input signal are carried out the down-sampled low-pass filtering of carrying out respectively earlier before.
As claim 6 or 7 described echo estimation approach, it is characterized in that 8, the described peaked method of echo of finding out of step 3) is: compare by each some amplitude and search out the echo maximum to cross correlation results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 02146714 CN1219391C (en) | 2002-11-04 | 2002-11-04 | Echo eliminator and method for evaluating echo |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 02146714 CN1219391C (en) | 2002-11-04 | 2002-11-04 | Echo eliminator and method for evaluating echo |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1499809A CN1499809A (en) | 2004-05-26 |
| CN1219391C true CN1219391C (en) | 2005-09-14 |
Family
ID=34232851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 02146714 Expired - Fee Related CN1219391C (en) | 2002-11-04 | 2002-11-04 | Echo eliminator and method for evaluating echo |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1219391C (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1842110B (en) * | 2005-03-28 | 2010-04-28 | 华为技术有限公司 | Echo eliminating device and method |
| CN101026659B (en) * | 2006-02-23 | 2010-04-07 | 中兴通讯股份有限公司 | Method for realizing echo time delay positioning |
| EP1845699B1 (en) * | 2006-04-13 | 2009-11-11 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Audio signal decorrelator |
| CN101272161B (en) * | 2008-04-30 | 2012-12-05 | 中山大学 | Self-adapting echo cancellation method |
| CN106571147B (en) * | 2016-11-13 | 2021-05-28 | 南京汉隆科技有限公司 | Method for suppressing acoustic echo of network telephone |
| CN108831491B (en) * | 2017-05-04 | 2020-07-17 | 展讯通信（上海）有限公司 | Echo delay estimation method and device, storage medium and electronic equipment |
| CN107071197B (en) * | 2017-05-16 | 2020-04-24 | 中山大学花都产业科技研究院 | Echo cancellation method and system based on full-phase multi-delay block frequency domain |
| CN107277209B (en) * | 2017-07-27 | 2020-05-29 | 维沃移动通信有限公司 | Call adjustment method and mobile terminal |
| CN109451398B (en) * | 2018-11-16 | 2021-03-19 | 珠海市杰理科技股份有限公司 | Acoustic feedback cancellation apparatus, acoustic feedback cancellation method, and audio processing system |
| CN111294474B (en) * | 2020-02-13 | 2021-04-16 | 杭州国芯科技股份有限公司 | Double-end call detection method |
| CN111246036A (en) * | 2020-02-17 | 2020-06-05 | 上海推乐信息技术服务有限公司 | Echo estimation method and device |
| CN112929055B (en) * | 2021-01-19 | 2022-12-27 | 北京集睿致远科技有限公司 | Echo cancellation circuit |
-
2002
- 2002-11-04 CN CN 02146714 patent/CN1219391C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1499809A (en) | 2004-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101017766B1 (en) | Spectral domain, non-linear echo cancellation method in a hands-free device | |
| CN1219391C (en) | Echo eliminator and method for evaluating echo | |
| US7002913B2 (en) | Packet loss compensation method using injection of spectrally shaped noise | |
| CN101026659B (en) | Method for realizing echo time delay positioning | |
| CN105825864B (en) | Both-end based on zero-crossing rate index is spoken detection and echo cancel method | |
| CN1111973C (en) | Echo cancelling system for digital telephony applications | |
| CN1139501A (en) | Method for determining location of echo in an echo cancellar | |
| CN104052526A (en) | Clipping Based On Cepstral Distance For Acoustic Echo Canceller | |
| CN106571147B (en) | Method for suppressing acoustic echo of network telephone | |
| CN1059628A (en) | Technology with proportional injection noise removing residual echo | |
| CN101179635B (en) | Device, method and system for performing echo control to hand-free telephone | |
| CN110995951B (en) | Echo cancellation method, device and system based on double-end sounding detection | |
| CN1650537A (en) | Echo cancellation | |
| CN1269926A (en) | Method and apparatus for echo estimation and suppression | |
| CN1192510C (en) | Method and apparatus for controlling echo on both sides of a connection | |
| CN1842110B (en) | Echo eliminating device and method | |
| CN1691716A (en) | Echo eliminator | |
| CN1350727A (en) | Pure delay estimation | |
| CN1592236A (en) | Method and device for testing speech quality | |
| CN107071197B (en) | Echo cancellation method and system based on full-phase multi-delay block frequency domain | |
| CN1275001A (en) | Adaptive filter emitting control in echo suppressor using structure lagging amplitude distribution estimation method | |
| PL216396B1 (en) | The manner and system of acoustic echo dampening in VoIP terminal | |
| FR2791492A1 (en) | ADAPTIVE IDENTIFICATION METHOD AND DEVICE, AND ADAPTIVE ECHO CANCER USING SUCH A METHOD | |
| CN1780165B (en) | Echo eliminator and elimination method | |
| CN1154242C (en) | Digital echo eliminator |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20050914 Termination date: 20181104 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |