similar to: Request for sample snippet of how to use jitter buffer

Hi Jean-Marc, Thanks for your response. Given a worst case scenario, what is the "worst case" latency (in terms of Speex frames) that the jitter buffer algorithm will incur? We're trying to determine the worst case hard number. Sorry for unclear question below; what I was trying to ask is that given a worst case latency (which I'm asking in the first question) inherent in

Thanks Jean-Marc. Is there a Speex trunk for the C64xx I can use? Or do I have to figure what options to turn on (in configure) when building the library? Thanks. Regards, Andy Ngo ----- Original Message ---- From: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> To: ndno72-speex@yahoo.com Cc: jtrantow@ieee.org; speex-dev@xiph.org Sent: Tuesday, December 5, 2006 2:52:02 PM Subject:

Tried your fixed_generic.h change but that didn't help. Andy ----- Original Message ---- From: Andy Ngo <ndno72-speex@yahoo.com> To: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> Cc: speex-dev@xiph.org Sent: Wednesday, December 12, 2007 4:13:35 PM Subject: Re: [Speex-dev] 4kbps sounds robotic on TMS320C64 Jean-Marc, Yes, fixed-point is enabled (#define FIXED_POINT in

Andy Ngo wrote: > Thanks Jean-Marc. Is there a Speex trunk for the C64xx I can use? Or > do I have to figure what options > to turn on (in configure) when building the library? Thanks. Support is in the main version. Just look at the TI/ directory for extra TI-related stuff. Jean-Marc > Regards, > Andy Ngo > > ----- Original Message ---- > From: Jean-Marc Valin

Daniel Schmidt a ?crit : > I found the cause of the problem. The function shift_timings can > produce overflows in the timing array if the jitter is huge or the > time units are very short. After changing the timing values' type from > spx_int16_t to spx_int32_t it seems to work. Hmm, I always assumed there wouldn't be any overflows. What parameter range are you using that

Hi, Is there a white paper describing how the jitter buffer in Speex works and what "side effects" (such as introducing delay) can happen? In the past, I have implemented a similar "dual buffer" algorithm to try to smooth out lost or delayed packets with a fixed know delay (no larger than 2 times the Speex frame). Thanks. Regards, Andy -------------- next part

Hi Jean-Marc, I have tested Speex in fixed-point mode on my PC without clicking noise. As I mentioned below, I tested Speex in fixed-point mode running natively on the ARM side of the DaVinci without click noise. I only get the clicking noise when running Speex on the DSP side. For the TI C64+ DSP on the DaVinci processor, the TI-specific switches doesn't do much other than defining the

Hi, I'm trying to use the jitter buffer feature that comes with Speex but I'm getting unexpected latency. I wrote a client application that does VOIP-like functions and without using jitter buffer, the end-to-end latency is around 250 ms (I'm using lowband 5.97 kpbs). However, when I tried to incorporate the jitter buffer feature, the latency would grow as time elapsed (up to a few

Hi, Our project is using the jitter buffer feature built in Speex. We noticed there are some latency when using the jitter buffer. Does anyone know what is the "worst case" latency inherent in the jitter buffer algorithm? I believe someone already mentioned that it's adaptive but is there a worst case hard number (in terms of 20ms Speex frames)? I'm not familiar with the

Hi, I looked at the jitter buffer code and it seems like the maximum number of frames that Speex can hold up to is 200 (SPEEX_JITTER_MAX_BUFFER_SIZE). This is equivalent to 4 seconds (20msec/frame * 200 frames). Can I just reduce this constant to limit the size? I know I'll reduce the smoothness of delayed frames but I want to reduce the delay in case my audio application isn't

Jerry, Thanks for the response; basically, I need to get just the narrow band working now (maybe use advanced features such as jitter buffer, echo cancellation, VAD, etc in the future); the DSP will do the encode and decode simultaneously as data are available. My main objective is just do a voice application that runs on the ARM side of the DaVinci with the codec running on the DSP side; the

Hi all: How to use jitter.c, I see this file is not include in the lib. Sorry for this newbie Q.

Sorry everyone, but I figured it out; it's working now. The problem was in my monoToStereo and stereoToMono functions (the Linux OSS driver on the ARM side only supports stereo so I had to convert it to mono before feeding it to Speex); also I had an alignment issue with my buffers I was using on the DSP side. Thanks Jim/Jean-Marc for your help! Regards, Andy ----- Original Message ----

Hi, I'd like to report a bug in speexclient/speex_jitter_buffer.c. In the function speex_jitter_get(), after the line: packet.data = data; add the line packet.len = 2048; If this variable is uninitialized, then in jitter_buffer_get() (in jitter.c), the predicate "jitter->packets[i].len > packet->len" is undetermined. When this happens, the wrong amount of

Hello Jean-Marc, thank you for your reply. > > I found the cause of the problem. The function shift_timings can > > produce overflows in the timing array if the jitter is huge or the > > time units are very short. After changing the timing values' type from > > spx_int16_t to spx_int32_t it seems to work. > Hmm, I always assumed there wouldn't be any overflows.

Hello Jean-Marc, Hello speex-dev-people, Jean-Marc Valin wrote: > All I can suggest is having a look at my simple client at: > http://people.xiph.org/~jm/speexclient/ Unfortunately, this does not compile: <<<<< snip >>>>> bjoern@phobos:~/beleg/speex/speexclient-0.0-dev$ ./configure checking for a BSD-compatible install... /usr/bin/install -c checking whether

Hmm, I tried... I completly understand an idea of jitter buffer and I know there is a lot of kinds of this solution (eg. AJB - Adaptive Jitter Buffer). I simply want to know what type is used in speex codec and how could I use that. What is the reason for using jitter buffer implemented in speex against to my own (implemented at lower layer - transmission layer - eg. rtp). Kapul On Tue, Sep

Hi, I have a SIP client which is registered to asterisk. Asterisk is registered to a SIP trunk and also handles the media. Now since my client has some issues in its RTP Tx, which seems to have some amount of jitter (mean jitter as per ethereal trace is about 17ms, max jitter is 20 ms and max delta is 85 ms), to over come that I have enabled jitter buffer in the SIP channel by setting sip.conf

Good question. I'm coming to the conclusion that using plain UDP and "home-grown" packet construction for transmitting the speex data (with timestamp/sequence counter) and implementing jitter control on the receiver end is an adequate implementation for a VoIP application. Assuming of course that I don't care about any interoperability issues with other applications etc. I was

Greetings list, Some time ago (probably about a year ago now) we disabled IAX jitter buffering on all our boxes because it was causing issues in a mixed 1.0 and 1.2 environment. One thing I've noticed over the last few months as more and more clients have moved from the 512k/1mb/2mb ADSL connections they were using onto "up to 8mb" connections is that whilst overall throughput is a