similar to: mdf -- better adaption of W?

>> Generate a test signal (10+x sine waves per frame), where x increases by >> one for each iteration, and wraps around at 100. > > Testing with sine waves is usually not a good idea. If you intend on > cancelling speech, then test with speech. Ok, I tested more extensively with both music and two-way speech. More on this below. >> However, when peeking at the

> Generate a test signal (10+x sine waves per frame), where x increases by > one for each iteration, and wraps around at 100. Testing with sine waves is usually not a good idea. If you intend on cancelling speech, then test with speech. > First off, it seems W is stored 'backwards'. The first values are for the > oldest frame, ok :) right. > However, when peeking at the

>> Actually, computing the "power spectrum" for each frame of W shows >> how large an ammount of the original signal at time offset j the >> echo canceller thinks should be removed from the current input frame. > > Careful when looking at W because of how the real and imaginary parts > are packed in the array. Err. Ok, as I got it, 'bin 0' has it's

> Actually, computing the "power spectrum" for each frame of W shows > how large an ammount of the original signal at time offset j the > echo canceller thinks should be removed from the current input frame. Careful when looking at W because of how the real and imaginary parts are packed in the array. > If you compute W*X for each j and ifft, you'll get the >

>> I'll need to add support for saving audio to my program, so I can give you >> the "actual" sampled loudspeaker and mic files, and I'll also need to get >> hold of a test person again. (I had a friend with a friend who has an >> exceptionally clear voice. My own "aaaaaa" is far too muddy to cause >> this). I'll try to get this done

> Err. Ok, as I got it, 'bin 0' has it's amplitude in W[0], bin 1 to N-1 has > it's real part in W[i*2-1] and it's imag in W[i*2], and finally the > nyquist amplitude is in W[N-1] Not quite, it's packet "real, real, imag, real, imag, ...". > I took this from how power_spectrum() computes, so I might be off :) But power_spectrum() handles that

Jean-Marc Valin <Jean-Marc.Valin@USherbrooke.ca> wrote: > > > Anyway, I found a few papers on multisource echo cancellation. Unless it's > > already a feature on the short-term "TODO" list, I'll take a stab at > > adding simple multisource cancelling after christmas. > > Multi-channel echo control is the subject of my post-doc and it will >

Hi, I wanted to play around with the echo canceller so I tried the testecho program. As speaker sound I fed it with a 440 Hz sine wave, and as mic with a 550 Hz + 440 Hz sine waves sound (but the 440 Hz component half the amplitude). I kind of expected to get a somewhat clean 550 Hz sine out, but it doesn't seem to work that way. Are there better samples I can use with testecho to see

Hi, there is an interesting case when the FLAC encoder (using 1.2.0) is given simple waves. Simple waves means: I have a list of {frequency, duration, pause} tuples that define the monophonic tune. In other words, exactly one frequency is played at a time. This is the original dataset from 1989 (driving a PC speaker back then): $ ls -l ihold.sd -rw-r--r-- 1 jengelh users 20616 Aug 14 00:57

Hello all, I am trying out the echo cancellation of Speex (1.2beta2) on Windows XP. I use the Windows Media API (i.e. waveInOpen, waveOutOpen etc) for playback and capture, but I cannot make echo cancellation works in my use case. Here is a description of what I tried: (the sound is in 8kHz and 16 bits per sample, the frame size is 240, tail_length is long enough to cover the far end to near end

My tuneR sawtooth wave function generator is broken. When I use the sine function, I get exactly what I expect: a sine wave whose frequency is defined by the freq parameter. In particular, higher frequencies have shorter wavelengths (more cycles per second means shorter waves). When I create a sawtooth wave, the opposite seems to occur: higher frequencies result in longer waves. But that?s not

On Nov 26 22:39:27, hans at stare.cz wrote: > ./test_replaygain.sh fails for me in tonegenerator(), saying: > > ./test_replaygain.sh[91]: mawk: not found > Testing FLAC replaygain 8000 (8000 x 1) ... -: ERROR: got partial sample > > Apparently, the tone-generating awk script does not work with > my system's awk, which is "awk version 20110810" as distributed

Ccing the maintainer if the tuneR package. Looks to me like sawtooth (and square) don't behave as expected when using xunit="samples". Workaround is to use xunit="time" instead: sawtooth(110,duration=1/100,samp.rate=sample_rate,xunit="time") I looked at the code but found it to be opaque. -- Sent from my phone. Please excuse my brevity. On November 12, 2017

Thank you for you quick reply, Jean-Marc. I have just used a human voice to replace the sine wave, but the result is like a broken voice after tens of seconds. Do you have some sample voice in .sw that I can test with? I want to have one set or a few sets of sample voice that work for my understanding and debugging. This is the output from echo_diagnostic.m in my human voice test: Far end to

Greetings, I have several sets of oscillation data and would like to estimate the parameters of a sine function to each set (and hopefully automate this). A colleague provided an excel sheet that uses solver to minimize the RSS after fitting the sine function to each data set, but this cumbersome and difficult to automate. Is there a method in R for fitting a given sine function to a

Hello, I am testing speex 1.1.6's echo canceller. I am using testecho.c, with a few modifications to get it to run on Windows. My problem is that I am unable to get the echo cancellation to work correctly. I am working on an audio conferencing software, and one issue we have is sometimes the microphone picks up what is being played through the headset, resulting in an echo of the other

On Aug 07 10:08:43, petrparizek2000 at yahoo.com wrote: > > What sine sweep exactly? > > An exponential sweep. It started slightly below 24 Hz and ended almost at 24 > kHz. And it was 50 seconds long. > > > How did you obtain it, > > I used Angelo Farina's "Aurora" modules. One of them is called "Generate > sine sweep". Can you please

I am having a hard time interpreting the phase and coherency plots. x is two timeseries that occur at the same time i.e. a b 1 11.2 12.3 16 11.3 12.4 31 11.4 12.5 46 11.5 12.6 ...etc even though my example is does not show this they are oscillating at more or less the same frequency just shifted by t=x (imagine two sine waves offset with the 2nd sine

> What sine sweep exactly? An exponential sweep. It started slightly below 24 Hz and ended almost at 24 kHz. And it was 50 seconds long. > How did you obtain it, I used Angelo Farina's "Aurora" modules. One of them is called "Generate sine sweep". > and how exactly did you encode and decode it? 1) Opusenc --bitrate 12 --downmix-mono Sweep50.wav

Guys (and girls), I\'ve been testing Ogg Vorbis quality at home, and I thought I\'d share the results with you. For those who want the conclusion, Vorbis 1.0 RC3 sounds better than LAME 3.70 vbr/cbr (which I\'m told now that it\'s broken for VBR), at less bitrate, however, it\'s not perfect (and probably never will, since it\'s a lossy audio codec) and could use some