similar to: Quality degradation on new versions

Hi Jim, First of all - thanks, turning the highpass filter off was what I needed, and the waveforms match now. But, when i did the PESQ tests again I found an interesting result : version 1.0.5 still got a slightly better average score, but the standard deviation on version 1.2 beta1 was much smaller. The cause for that is this - on some samples versions 1.0.5 and 1.2beta2 produced a single

Hi Aviv, Does the audio sound bad? You can try turning off the highpass filter (which was not in 1.0.5). This code is from ti/testenc-TI-C5x.c in the source tree: /* Turn this off if you want to measure SNR (on by default) */ tmp=0; speex_encoder_ctl(st, SPEEX_SET_HIGHPASS, &tmp); speex_decoder_ctl(dec, SPEEX_SET_HIGHPASS, &tmp); - Jim ----- Original Message ----- From:

Dr. Williams, I ran across your inquiry on one of the R-help mailing lists regarding digital filter design and implementation. I found no response to your email in the archives and was wondering if you were able to find anything. Thanks, Israel -- Israel Christie, Ph.D. Email: ichristie at gmail.com Phone: 865.766.0214 Mobile: 865.406.4615

Hi, OK, basically it appears to be an overflow of the decoded signal, but I can't reproduce it with speexenc/speexdec. There's an overflow occurring instead of what should be saturation. Is it possible you're using speex_decode() and converting from float to short without first checking for overflows? Normally speex_decode_int() should do it for you, but if you use the float version

Hello, I just signed up to this mailing-list (note: my first mailing list at all), because I'm having some problems related to speex. Let me just introduce you to what I'm doing. I am writing a short (really short) paper about VoIP techniques, especially audio codecs for speech. I pointed out basic technologies behind audio codecs; vector quantization, lpc, long-term prediction and some

Greetings everyone, I was about to compare AEC performance between 1.1.12 and 1.2beta1 when I noticed something. If I configure (and compile) speex-1.1.12 with ./configure --enable-shared=no --enable-static=yes it compiles and works as expected: I can run a mic and speaker signal through testecho, it runs in a reasonable amount of time (about 23 secs for 3 minutes of audio) and I get back

Francois Menard (Mailing List Account) wrote: > > did you try speex in wideband mode ... what bitrate? > > f. 15kbps mode. It does significantly better with a different speaker (the male does better than the female), as well. I'm considering purchasing a commercial PESQ license and wrap the PESQ software in a server application. I could then allow other folks to run PESQ

Hello, I have this rare and weird problem in the encoder : when I encode and decode this specific speech sample (Attached) the first frame comes out wrong. I found that the error is in the encoder. My settings are : Version 1.2beta2, on linux, in floating point. Wideband. The highpass filter turned off. Perceptual enhancment is turned off. Quality is 9. The weird thing about it is that if I use

I'll try to make some shorter samples later, but for now here are the ones I have tried with: http://www.surfnonstop.com/~bandit/speex/1.2beta1_AEC_garble/ The original recordings are in mic.raw and spk.raw. Jean-Marc Valin wrote: > You may have triggered an instability problem. Can you upload your files > somewhere so I can have a look at them? > > Jean-Marc > > Andras

Hello! Some time back, I added the Speex protocol to my version of VOCAL (www.vovida.org, VOIP tool). Recently, I also added PESQ (automated voice quality testing algorithm) to my tool and have been running some tests on a clean network. The source file is a woman reading some phrases meant to test various aspects of codecs... Speex has a respectable result of 3.67 Some other codecs I've

Hi, I am using 1.2beta2 on a C5416 doing narrowband in the decode function I am having a problem with the execution of my program vectoring off to random interrupts that are not used. I have tracked down the error to the highpass function called in nb_decode -very close to the end of the function-context: it is called here: ln 1719 for me if (st->highpass_enabled) highpass(out,

Hi Kevin, Are you saying that the quality is good at 20 ms and bad at 10 ms, or the reverse? Also, is this speech or music? What tool, what options? In general, it helps a lot if you post the sample (input and output). Cheers, Jean-Marc On 06/03/2016 12:48 PM, Kevin Connor wrote: > Hi Opus list, > > I'm noticing a discontinuity in the quality between use of 10ms and > 20ms

Jean-Marc Valin wrote: >> What commercial VOIP test products can I turn to in order to objectively >> evaluate your CODEC; in order to analyze audio streams for MOS scores and >> performance parameters? >> > > The MOS evaluation is subjective, not objective and actually involves > getting (many) real people to listen to the audio, not just buying a >

Hello, Is there any way I can use the gregexpr functions (or a different function) in a manner that will also return overlapping (i.e. non disjoint) regular expressions? For instance, when running gregexpr("AAA","AAAAAA"), I get two matches, one at position 1 and one at position 4. I'd like to receive 4 matches at positions 1, 2, 3 and 4. Thanks, Schraga

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I've been trying to get the timidity system running as a daemon. I wrote the following init script: #!/bin/sh # # timidity # ### BEGIN INIT INFO # Provides: timidity # Required-Start: # Required-Stop: # Default-Start: 2 3 4 5 # Default-Stop: 0 1 6 # Short-Description: Add and remove timidity # Description: ### END INIT INFO .

Hi Jim, > Build 11169 in SVN works correctly. Good. I'll try not to forget the EXTEND32 from now on. > I have attached a zip file (renamed > .txt) with a patch to bits.c to make the byteswapping for TI DSPs > consistent. Seems like unzip can't read it. Either it's in an unknown format or the file got corrupted. Could simply send as multiple (uncompressed)

A bit of a perennial I'm afraid. My wife has filled up her iPhone 6 with photos, and wants them moved onto my computer. I'm running CentOS 7 patched about 30 minutes ago. Needless to say the computer can't see the data on the iPhone, though it does recognise the phone as an iPhone. Any suggestions (well any that don't involve a steam roller, sledge hammer or GBH to the whole of

Hi, I was just trying to use speex for sampling frequency >48KHz. In the original Speex-1.0.5 its restricted only upto 48KHz. I tired to modify it by changing the boundary conditions( the error conditions, i.e. if sampling freq >48KHz, it gives error) in /src/speexenc.c and then it atleast doesnt give the error, there is flow in decoding or encoding(i think). I suspect there are other

I am specifically using Celt Wideband (48kHz) over WiFi multicast that naturally leads to lost packets and am trying to minimize the impact to the audio. I implemented PLC but the audio it produces is robotic. Have I implemented PLC correctly? Checking the waveform it is using the previous received waveform to fill in a missing packet but not the full waveform so it has to repeat. Basically,

Hi, I am wondering if you may be able to help me. I have an application that has been developed by a third party and is being used in some field locations. My challenge is that I have no way of independently testing the product across the WAN at this point. All testing they have proposed required manual intervention and subjective analysis. I am not comfortable deploying the solution in this