eden-emu/opus
18
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
247 commits 285 branches 59 tags 26 MiB
Commit graph

247 commits

This branch
This branch
All branches
Author SHA1 Message Date
Jean-Marc Valin
45018cbfa2 Doing stereo mixing adaptively based on amplitude difference. 2008-01-11 14:02:07 +11:00
Jean-Marc Valin
a2482aceb4 Conversion to 18-band split and fixed a few related bugs. 2008-01-11 11:24:22 +11:00
Jean-Marc Valin
43e224817c oops, trying not to overflow the 64-bit max allocation. 2008-01-10 17:26:30 +11:00
Jean-Marc Valin
f8dda66a8e Tuning of the stereo mode, including conversion to the 18-band split. 2008-01-10 17:02:36 +11:00
Jean-Marc Valin
a4833ffada Stereo decoding working again (fixed a few issues in the encoder at the same 
time)
 2008-01-10 15:34:00 +11:00
Jean-Marc Valin
021478e252 More stereo work (doing mid-side quantisation of the energy) 2008-01-09 15:31:43 +11:00
Jean-Marc Valin
fdca84b42f Some stereo work (breaks the decoder for now) 2008-01-09 10:44:18 +11:00
Jean-Marc Valin
95c59ea647 Updated the 18-band mode (mode0) to use the low-overlap MDCT 2008-01-08 16:22:37 +11:00
Jean-Marc Valin
2c53c14904 Oops, forgot to add the VQ table! 2008-01-08 11:29:37 +11:00
Jean-Marc Valin
b21b2a404e Switched to mdct mask computation 2008-01-07 23:04:43 +11:00
Jean-Marc Valin
0a3e79aa11 Pseudo-masking curve for the MDCT 2008-01-07 13:36:01 +11:00
Jean-Marc Valin
825ec60b53 Completed low-overlap MDCT support by reducing the synthesis 
latency as well.
 2008-01-05 22:49:23 +11:00
Jean-Marc Valin
2875f6bf8d Now using 8 bands for the pitch gain, with a 128-entry codebook. 2008-01-05 13:32:32 +11:00
Jean-Marc Valin
a5431bfb09 Allowing reduced overlap MDCTs as an alternative 
to multiple MDCTs per frame. Not complete yet.
 2008-01-03 20:53:01 +11:00
Timothy B. Terriberry
71c9bbff02 Updated pulse coding to simpler (slightly faster) code included with 
http://people.xiph.org/~tterribe/notes/cwrs.html
Removed dead code.
 2008-01-01 07:18:06 +11:00
Jean-Marc Valin
26c9e1452a Comments/cleanup, no code change 2007-12-31 21:27:54 +11:00
Jean-Marc Valin
0d227d86e5 Got the intra-band prediction/copy to work correctly with 
pulse spreading (and to work at all).
 2007-12-31 16:12:12 +11:00
Jean-Marc Valin
97252d0365 Spreading function inside quant_bands() 2007-12-30 20:39:14 +11:00
Jean-Marc Valin
e84962bcd9 Doing multiple rotation passes instead of one produces better pulse spreading. 2007-12-30 00:17:44 +11:00
Jean-Marc Valin
571e861dc3 more exp_rotation() cleanup 2007-12-30 00:06:10 +11:00
Jean-Marc Valin
559e8ff602 simplified exp_rotation() 2007-12-30 00:02:00 +11:00
Jean-Marc Valin
6c24524b88 Added a DCT in time direction when multiple MDCTs are used within the same 
frame.
 2007-12-29 23:27:42 +11:00
Jean-Marc Valin
46014ca49a Propagating perceptual weighting around (not used yet). 2007-12-14 13:47:04 +11:00
Jean-Marc Valin
36d78e6f74 Band energy quantisation info now in the mode directly 2007-12-13 14:59:30 +11:00
Jean-Marc Valin
4a897680e3 Working on some stability issues (appears to be solved by making the pitch 
projection less aggressive). Also, fixed a 64-bit overflow in the stereo mode
and added a "band rotation" function.
 2007-12-12 00:45:15 +11:00
Jean-Marc Valin
f347dd3b3d Fixed parallel build 2007-12-11 18:01:22 +11:00
Jean-Marc Valin
f8db800c44 Added support for codebooks up to 64 bits. 2007-12-11 14:52:56 +11:00
Timothy B. Terriberry
f13fea7b53 Added ec_{enc|dec}_bits64 and ec_{enc|dec}_bits64. 
Updated broken paper URLs.
Cleaned up non-multiply-free range coder documentation (e.g., stop claiming
it's multiply-free).
 2007-12-11 13:25:57 +11:00
Timothy B. Terriberry
d710177d0a Documentation fixes and a couple of other minor edits. 2007-12-11 13:25:00 +11:00
Jean-Marc Valin
9d35ccdaea Fixed stereo version of the pitch estimator 2007-12-10 17:57:19 +11:00
Jean-Marc Valin
bf94045f41 Stereo almost works now (still need to fix the pitch estimate and add more 
pulses)
 2007-12-10 17:10:29 +11:00
Jean-Marc Valin
ffa13474ef Adding stereo support. Doesn't work yet, but at least it doesn't break 
mono
 2007-12-10 16:54:17 +11:00
Jean-Marc Valin
14bccef8e7 Comments on the spreading function 2007-12-10 13:13:58 +11:00
Jean-Marc Valin
991410be59 Added a simple psychoacoustic model (still needs a lot of work). 2007-12-10 01:28:34 +11:00
Jean-Marc Valin
aaab1698b4 Fixed a stupid error caused by kdevelop in the configure.ac 2007-12-09 13:02:25 +11:00
Jean-Marc Valin
11f01729e6 Some cleaning up, a few less warnings and the decoder no longer does an 
infinitete loop in Laplace decoding when the data is corrupted.
 2007-12-09 01:19:36 +11:00
Jean-Marc Valin
06ee7f9083 Updated README 2007-12-09 00:55:49 +11:00
Jean-Marc Valin
1677aa9caa code cleanup: all inverse MDCTs in the same function 2007-12-08 01:13:34 +11:00
Jean-Marc Valin
fb83fb422f Very basic packet loss concealment code 2007-12-08 00:56:04 +11:00
Jean-Marc Valin
aa5bd61cb7 Another experimental mode with longer frames. 2007-12-08 00:27:10 +11:00
Jean-Marc Valin
6baa8e18d9 A bit of quality tuning, plus created a higher quality mode 2007-12-08 00:08:37 +11:00
Jean-Marc Valin
98d2a4917e No more cheating, everything fully quantised 2007-12-07 22:46:47 +11:00
Jean-Marc Valin
9c3e22c440 Moving to non-multiply-free entropy coder 2007-12-07 22:25:31 +11:00
Jean-Marc Valin
fc43dbb7d6 A bit of probability tuning to reduce bit-rate variations 2007-12-07 19:12:51 +11:00
Jean-Marc Valin
b7ceec2bae oops, forgot laplace.h 2007-12-07 18:54:59 +11:00
Jean-Marc Valin
0c79f09e14 debug code 2007-12-07 18:02:56 +11:00
Jean-Marc Valin
c57ac95f17 Actually enabled the probability model 2007-12-07 17:58:44 +11:00
Jean-Marc Valin
143fbc3138 Decent starting point for energy probabilities 2007-12-07 17:48:18 +11:00
Jean-Marc Valin
e38c7c18f7 Energy is now Laplace-encoded (very poorly for now) 2007-12-07 17:10:14 +11:00
Jean-Marc Valin
8143be3026 energy decoding partially done (cheating a bit) 2007-12-07 16:40:39 +11:00