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 |
|
Jean-Marc Valin
|
b726185d5b
|
Fixed Laplace encoder
|
2007-12-07 16:06:27 +11:00 |
|
Jean-Marc Valin
|
671bf403e6
|
More work on decoding (still cheating)
|
2007-12-07 15:36:39 +11:00 |
|
Jean-Marc Valin
|
0aa3903a4a
|
Cheating decoder now produces the same result as the encoder
|
2007-12-07 15:09:58 +11:00 |
|
Jean-Marc Valin
|
6e9058adb9
|
Nearly working cheating decoder.
|
2007-12-07 14:59:06 +11:00 |
|
Jean-Marc Valin
|
70c8ffdd55
|
More decoding work
|
2007-12-07 14:20:01 +11:00 |
|
Jean-Marc Valin
|
fc08d0a6d6
|
Algebraic codebook decoding (not tested yet)
|
2007-12-07 13:26:15 +11:00 |
|
Jean-Marc Valin
|
269d40a5c0
|
Added decoder code (not working yet)
|
2007-12-07 11:29:45 +11:00 |
|
Jean-Marc Valin
|
bb57fe44e2
|
Most of the Laplace encoding works, except that ec_decode() is still not giving
me what I expect
|
2007-12-07 10:56:25 +11:00 |
|
Jean-Marc Valin
|
468c5395b4
|
More code for laplace decoding (doesn't work)
|
2007-12-07 09:50:22 +11:00 |
|
Jean-Marc Valin
|
6887fb5181
|
Code for encoding Laplace-distributed variables (doesn't work).
|
2007-12-07 01:40:02 +11:00 |
|
Jean-Marc Valin
|
c245a226ce
|
minor cleanup
|
2007-12-06 19:14:20 +11:00 |
|
Jean-Marc Valin
|
9cace64e5f
|
Fixed codebook entirely quantised
|
2007-12-06 17:44:09 +11:00 |
|
Jean-Marc Valin
|
e8376605d5
|
proper COPYING file
|
2007-12-06 17:33:10 +11:00 |
|
Jean-Marc Valin
|
e5bab14e4f
|
Doing init/reset of the entropy coder properly
|
2007-12-06 16:51:19 +11:00 |
|
Jean-Marc Valin
|
29ccab8aac
|
Some work on index packing
|
2007-12-06 15:39:38 +11:00 |
|
Jean-Marc Valin
|
bc5cedf26f
|
Linking with libentcode
|
2007-12-06 15:18:25 +11:00 |
|
Timothy B. Terriberry
|
2ec8d9e522
|
Multiplier-free entropy coder
|
2007-12-06 15:09:53 +11:00 |
|
Jean-Marc Valin
|
aeb4467d7f
|
improved energy quantisation/prediction
|
2007-12-06 13:50:14 +11:00 |
|
Jean-Marc Valin
|
06311335b5
|
More energy quantisation work
|
2007-12-06 11:13:56 +11:00 |
|
Jean-Marc Valin
|
8b0137aa42
|
Quantisation of band energies (adding files)
|
2007-12-06 07:51:53 +11:00 |
|
Jean-Marc Valin
|
fa248735ea
|
Quantisation of energies
|
2007-12-06 07:51:27 +11:00 |
|
Jean-Marc Valin
|
96870d9334
|
conversion to modes complete
|
2007-12-05 21:14:22 +11:00 |
|
Jean-Marc Valin
|
3337e5d3e5
|
More modes work
|
2007-12-05 21:00:03 +11:00 |
|
Jean-Marc Valin
|
73e51b3e94
|
Converting the code to use the modes instead of global arrays.
|
2007-12-05 17:48:24 +11:00 |
|
Jean-Marc Valin
|
ecb36a3323
|
Adding mode infrastructure (still incomplete)
|
2007-12-05 01:31:49 +11:00 |
|
Jean-Marc Valin
|
1e7087e4cf
|
Compressing the innovation along the pitch direction
|
2007-12-04 13:05:43 +11:00 |
|
Jean-Marc Valin
|
25298f296a
|
Allowing to add pulses on top of intra-band prediction
|
2007-12-03 15:24:11 +11:00 |
|
Jean-Marc Valin
|
50d1116883
|
Fixed intra-frame prediction (added sign)
|
2007-12-03 14:34:52 +11:00 |
|
Jean-Marc Valin
|
4841a0a02b
|
Intra-frame prediction
|
2007-12-03 13:54:30 +11:00 |
|
Timothy B. Terriberry
|
c4541ae786
|
Code for converting back and forth between pulse vectors and the correspondig
VQ index.
|
2007-12-03 11:51:29 +11:00 |
|
Jean-Marc Valin
|
7f4701383b
|
Nothing to see here.
|
2007-12-02 20:55:22 +11:00 |
|
Jean-Marc Valin
|
f3efa3e132
|
Some improvements to the pitch period estimation
|
2007-12-01 01:55:17 +11:00 |
|
Jean-Marc Valin
|
6f7e83d8f4
|
Pre-emphasis, plus a few minor tweaks
|
2007-12-01 00:36:41 +11:00 |
|
Jean-Marc Valin
|
41af421540
|
Oops, forgot to add the Ghost/ceft vq files
|
2007-11-30 18:35:37 +11:00 |
|
Jean-Marc Valin
|
3c453121d2
|
Vector quantisation of the residual (copied from Ghost/ceft)
|
2007-11-30 17:41:09 +11:00 |
|
Jean-Marc Valin
|
dff7d36a88
|
Some pitch prediction work.
|
2007-11-30 16:57:43 +11:00 |
|
Jean-Marc Valin
|
991c0f0254
|
Code for computing band energies and normalising: adapted from CEFT.
|
2007-11-30 16:07:46 +11:00 |
|
Jean-Marc Valin
|
da72188b18
|
Forward MDCTs in a function
|
2007-11-30 15:17:42 +11:00 |
|
Jean-Marc Valin
|
4bf5fd8824
|
Pitch cleanup/bugfix
|
2007-11-30 15:01:48 +11:00 |
|
Jean-Marc Valin
|
14191b3ccd
|
Added pitch analysis. Doesn't crash, but otherwise untested.
|
2007-11-30 12:15:49 +11:00 |
|
Jean-Marc Valin
|
013c31d6e6
|
Got MDCT analysis-synthesis to work
|
2007-11-30 11:36:46 +11:00 |
|
Jean-Marc Valin
|
f02ba11950
|
MDCT analysis-synthesis (untested)
|
2007-11-30 01:10:42 +11:00 |
|
Jean-Marc Valin
|
8ebd345430
|
initial celt struct
|
2007-11-29 20:17:32 +11:00 |
|
Jean-Marc Valin
|
07ec79072d
|
Adding fields to the struct
|
2007-11-29 18:06:21 +11:00 |
|
Jean-Marc Valin
|
405025dea8
|
kdevelop project
|
2007-11-29 17:29:29 +11:00 |
|
Jean-Marc Valin
|
842b445e1b
|
There's nothing, but it now compiles
|
2007-11-29 17:19:47 +11:00 |
|