search for: nlms

Hi all, I have a query. I developed NLMS acoustic echo cancellation algorithm for real time voip application and using general purpose processor. Used more filter length (4096/2048) for NLMS and faced CPU consumption.With less filter length unable to estimate the exact echo signal and the quality of the speech also getting degraded. Cou...

Hello Jean-Marc Valin and all the Speex-dev team, Today I have a little question about patents and the AEC. I' am not a lawyer, but after a deep research in esp@cenet.com, I found that in August 2003, an AEC called "Double talk insensitive NLMS algorithm" (DNLMS) have been patented by Gary Q Jin, Gordon J Reesor, and others (US 6.608.897 B1). It uses a continuous adaptation step as speex, to avoid adaptatif filter divergence during double-talk. Is the computation method really the same as speex AEC (hope not) ? Somebody who have a...

...probably designed primarily for small communication devices with lower sampling rates and less demand for low latency. Nontheless, I must say that speex was one library that actually worked for me and was effective in echo/feedback cancellation! I tried to use some other adaptive filters (LMS and NLMS) the same way I am using speex, but they didn't appear to do what I need. I was wondering therefore, what type of adaptive filtering does speex use internally for echo cancellation? Does the processing consist of more than only what is traditionally called an adaptive filter? Best regards, Ja...

...setting echo tail length of AEC, but the output e(n) was not an echo canceled verson of x(n), It seemed that the ^h(n) never converged to the real impulse response h(n), why? and I didn't suspect the input files because the same x(n) and y(n) input could work well in another simple time-demon nlms AEC method which was implemented in malab. ???????? best regards, ???????? Johnson Peng ????????p_yuanjiang@sina.com ??????????2006-09-29

...117367 Dec 3 00:47 rsync-netware-nlm-2.5.5.tar.gz This is a pair of Novell Netware NLM compiled-binary executables for the rsync utility (the main daemon and a semaphore-based rsync-service stopping utility), with a sample configuration file, welcome-text file, and scripts to run the two NLMs. This is a port to NetWare of rsync v. 2.5.5, provided by Novell South Africa to a customer, who e-mailed them to me. He says they are included on Novell's Nterprise Branch Office CD #2. I believe all the code to be lawfully redistributable and subject to the GNU GPL. I have no idea...

...n the NES. I haven't tried these cases on 1.1.9. I'm wondering if some one can point me to what type of echo canceller speex is using? Or even to a description of the algorithm itself? I looked at the code, but didn't recognize the algorithm. (I could only tell it wasn't FIR or NLMS.) BTW, I'm running this on WinXP w/ VC++ 6.0. I had to cast things to float in lots of places (to get rid of warnings) and define away "restrict", but otherwise didn't have to change anything. - Eric Scott

...ength of AEC, but the output e(n) was not an echo canceled > verson of x(n), It seemed that the ^h(n) never converged to the real > impulse response h(n), why? and I didn't suspect the input files > because the same x(n) and y(n) input could work well in another > simple time-demon nlms AEC method which was implemented in malab. > > > ???????? > > best regards, > > ???????? Johnson Peng ????????p_yuanjiang@sina.com ??????????2006- > 09-29 > > > ------------------------------------------------------------------------ > > > ______...