similar to: fixed point conversion volunteer (OMG what am I doing?)

I'm willing to contribute some time to converting speex to fixed point. I'm not doing this for a particular product and I will receive nothing but some warm fuzzy feelings for "giving back to the community". Consequently, I don't want to do it all myself. So ... is anyone else currently working on this? Splitting the work into chunks should allow more people to write code

Hello Mark: Count me in. I'm an excellent C programmer and I've done assembly coding. Although I've not done float to int conversions, I'll pick it up quickly ... send me an example, please. Also, I direct The Tech Startup Connection here in Silicon Valley, so I have a very large resource pool. Check out

I submitted a version of testenc.c to this list MONTHS ago that fixed this problem. It uses a reconfigurable group delay. If you had given me cvs access I could've committed the changes myself. Anyway, the file is attached (again). Maybe this time you'll actually notice the email. Sorry if I seem a little short. It's just very annoying to put the time into learning and improving a

I thought I should be getting 20 words because in the windows app the command line repeats "nbBytes = 20" through the entire decode phase. I wonder how the frames got misaligned...right now I get 10 words of data followed by 10 words of 0's which I look at through a memory window in code composer. now I am pretty sure there is a terminator word that gets in there so that would make

I have no problems compiling with Borland compiler (in fact it helped uncover a couple of math issues during run-time). I use BDS 2006. What's happening in your case is that the Speex itself is not linked in. I suggest you create a library out of the C sources and link it in with your test app. HTH, .a -----Original Message----- From: speex-dev-bounces@xiph.org

Hello, I want to compile testenc.c with Borland C/C++ compiler, but get a lot of errors. What's wrong? Borland C++ 5.5.1 for Win32 Copyright (c) 1993, 2000 Borland testenc.c: Turbo Incremental Link 5.00 Copyright (c) 1997, 2000 Borland Error: Unresolved external '_speex_encoder_init' referenced from D:\SPEEX\TESTENC.OBJ Error: Unresolved external '_speex_decoder_init'

Had an appointment for these schmoes to come out and install another line. Was supposed to be 8-12. Its now 6PM and not even call. Missed 3 sales calls waiting on these jerks. No wonder customers were jumping ship to Vonage. -- Warm Regards, Lee

Hi, I've just found something really interesting and OMG ... it's amazing, just take a look <http://ndikynkiky.vendercamara.com/aexfbue?nxc> Nouveau -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://lists.freedesktop.org/archives/nouveau/attachments/20160520/12543c48/attachment-0001.html>

http://www.eweek.com/article2/0,1759,1591131,00.asp bkw -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.digium.com/pipermail/asterisk-users/attachments/20040514/165456d3/attachment.htm

http://blogs.computerworld.com/15082/omg_microsoft_patents_sudo_linux_and_macos_dead?source=CTWNLE_nlt_dailyam_2009-11-12 -- thanks ./francis -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.centos.org/pipermail/centos/attachments/20091112/1061174b/attachment-0003.html>

Mike, Generally "Invalid mode encounterd" == "frames are misaligned" You should be getting 20 bytes from the encoder each time, and passing 20 bytes to the decoder each time. Is it correct that you have modeled your main loop after testenc-TI-c5x.c? If you look at encoded silence with a binary editor, you should be able to see the 20-byte repetition pattern. You can also

DEAR: I SEEM A BUG FOR speex-1.2rc1.tar.gz ON BUILD IN USING IT. PLEASE LOOK: ===================================================================== make all-recursive make[1]: Entering directory `/tmp/speex-1.2rc1' Making all in libspeex make[2]: Entering directory `/tmp/speex-1.2rc1/libspeex' /bin/sh ../libtool --mode=link gcc -g -O2 -fvisibility=hidden -O3 -msse -o testenc

Actually, the SNR calculation in testenc has been broken for a while. The reason for taking the last frame was that the codec would introduce a one-frame latency. I changed that to half a frame (10 ms) a while ago and never updated testenc. Jean-Marc Le lun 01/09/2003 à 13:12, Mike Dunn a écrit : > Hi all, > > I'm new to the group. I'm looking at the speex code with an

Hi Jim, Actually, I don't see anything wrong with the internal structure having a different type than the interface, as long both types are big enough to hold the possible values (in this case 0 and 1). Though, as you pointed out, testenc needs to be fixed to use spx_int32_t instead of int. I'll change that. Jean-Marc Jim Crichton a ?crit : > st->highpass_enabled is typed

Hi, No, it only happens with speex_wb_mode. But like I said, it is because in wideband mode, the function sb_encoder_init is called and this function initializes the sampling rate using speex_*de*coder_ctl instead of speex_*en*coder_ctl In narrowband mode, the sampling rate is just initialized to 8000 st->sampling_rate=8000; BTW, I saw that my email was added to an existing thread instead of

Jean-Marc, The trouble is that one cannot identify the type difference without reading the nb_celp.c source file, and this parameter is treated differently than other, similar parameters. It would seem like the most intuitive approach would be to match the interface to the structure. This is a trivial point, of course. I have attached a patch to the two TI interface files, which adds the

Jean-Marc, Here are the patches, broken out one per file. I did some additional work in the testenc files based on Peter Mlakar's post to make it easier (though still a bit klunky) to try different rates. These changes also incorporate the file window.c into the builds, change the decoder startup delay to account for the change in the algorithm delay from 10ms to 5ms, and change bits.c

Hi Tom, Thanks for your reply, I'm having a few difficulty's following you advise. You mention that I should be calling 'speex_encoder_int' & 'speex_decoder_int', I'm having problems finding these functions in the speex header. I am currenlty calling speex_encoder_init & speex_decoder_init in my test code, I have repeated the relevent bits of the example

Hi, Do you get that problem with speex_nb_mode too (or only speex_wb_mode)? Jean-Marc Le lun 29/03/2004 à 05:35, Chris Flerackers a écrit : > Hi, > > I'm using speex 1.1.4 and when I query SPEEX_GET_SAMPLING_RATE > as in > speex_encoder_ctl ( m_pEncoderState, SPEEX_GET_SAMPLING_RATE, > &m_SampleRate ); > I get > 2 > The encoder is initialized with

For clarity I attached modified testenc.c I used to test my code. On 2/14/08, Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> wrote: > > I also think that having rame_num.c in libspeex, code size could be > > reduced and simplified slightly by using speex_skip_wb_frame() in > > nb_decode() instead of duplicating code. > > Well, the code is a bit different (e.g.