| .. |
|
_kiss_fft_guts.h
|
Converted a few double-precision constants to single precision
|
2010-04-16 20:57:56 -04:00 |
|
arch.h
|
Converted a few double-precision constants to single precision
|
2010-04-16 20:57:56 -04:00 |
|
bands.c
|
Comments, code cleanup
|
2010-06-19 10:27:28 -04:00 |
|
bands.h
|
Adaptive time-frequency resolution
|
2010-05-27 23:56:53 -04:00 |
|
c64_fft.c
|
Changed all the celt*int*_t types to remove the _t suffix, which is reserved
|
2009-10-17 14:35:13 -04:00 |
|
c64_fft.h
|
Changed all the celt*int*_t types to remove the _t suffix, which is reserved
|
2009-10-17 14:35:13 -04:00 |
|
celt.c
|
Removing old PCL, added plc.h
|
2010-06-20 23:01:33 -04:00 |
|
celt.h
|
Implemented CELT_SET_START_BAND
|
2010-05-19 13:38:36 -04:00 |
|
celt_header.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
celt_types.h
|
Changed all the celt*int*_t types to remove the _t suffix, which is reserved
|
2009-10-17 14:35:13 -04:00 |
|
cwrs.c
|
Removing the rest of the >32-bit PVQ code
|
2010-05-24 09:05:36 -04:00 |
|
cwrs.h
|
Removing the rest of the >32-bit PVQ code
|
2010-05-24 09:05:36 -04:00 |
|
dump_modes.c
|
shortWindow hadn't been used for a while
|
2010-05-07 21:20:21 -04:00 |
|
ecintrin.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
entcode.c
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
entcode.h
|
Removing unused code in the entropy coder
|
2009-10-23 21:44:54 -04:00 |
|
entdec.c
|
Provide direct implementations ec_{enc|dec}_bit_prob() that do not require a division instead of using the normal entropy coder path. This should be exactly equivalent to the existing code.
|
2010-05-29 23:45:18 -04:00 |
|
entdec.h
|
Change ec_{enc|dec}_bit_prob to take probabilities in units of 1/65536 instead of 1/256. This allows them to use a single 16x16->32 multiply instead of a 24x8->32 multiply. Also change the time-frequency resolution flag coding to ensure that "0" is always the most-probable symbol (i.e., that prob("1")<50%), as that's where all the rounding error accumulates.
|
2010-05-29 23:45:28 -04:00 |
|
entenc.c
|
Provide direct implementations ec_{enc|dec}_bit_prob() that do not require a division instead of using the normal entropy coder path. This should be exactly equivalent to the existing code.
|
2010-05-29 23:45:18 -04:00 |
|
entenc.h
|
Change ec_{enc|dec}_bit_prob to take probabilities in units of 1/65536 instead of 1/256. This allows them to use a single 16x16->32 multiply instead of a 24x8->32 multiply. Also change the time-frequency resolution flag coding to ensure that "0" is always the most-probable symbol (i.e., that prob("1")<50%), as that's where all the rounding error accumulates.
|
2010-05-29 23:45:28 -04:00 |
|
fixed_c5x.h
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
fixed_c6x.h
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
fixed_debug.h
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
fixed_generic.h
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
float_cast.h
|
Preparing for 0.5.0
|
2008-10-09 07:29:39 -04:00 |
|
header.c
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
kfft_double.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
kfft_single.c
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
kfft_single.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
kiss_fft.c
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
kiss_fft.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
kiss_fftr.c
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
kiss_fftr.h
|
License for the kiss-fft headers
|
2009-02-03 23:27:49 -05:00 |
|
laplace.c
|
Added a SMALL_FOOTPRINT option
|
2009-10-24 11:44:07 -04:00 |
|
laplace.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
Makefile.am
|
Removing old PCL, added plc.h
|
2010-06-20 23:01:33 -04:00 |
|
match-test.sh
|
Simplified the folding normalisation by creating renormalise_vector().
|
2008-07-20 23:14:31 -04:00 |
|
mathops.h
|
removing some warnings
|
2010-06-21 01:01:17 -04:00 |
|
mdct.c
|
Shrinking the MDCT's table by separately rotating by the 1/(8N) factor
|
2010-05-07 20:57:48 -04:00 |
|
mdct.h
|
Renamed MDCT functions to avoid symbol clashes with Vorbis
|
2009-11-21 21:57:44 -05:00 |
|
mfrngcod.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
mfrngdec.c
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
mfrngenc.c
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
modes.c
|
Remove unnecessary free of global stack on error
|
2010-06-08 21:01:41 -04:00 |
|
modes.h
|
Completely removed the old allocation table.
|
2010-06-03 23:29:35 -04:00 |
|
os_support.h
|
Misc changes for 0.7.1.
|
2010-01-16 19:12:06 -05:00 |
|
pitch.c
|
More work on variable frame size (getting rid of FRAMESIZE() )
|
2010-05-07 21:20:22 -04:00 |
|
pitch.h
|
More work on variable frame size (getting rid of FRAMESIZE() )
|
2010-05-07 21:20:22 -04:00 |
|
plc.c
|
Removing old PCL, added plc.h
|
2010-06-20 23:01:33 -04:00 |
|
plc.h
|
Removing old PCL, added plc.h
|
2010-06-20 23:01:33 -04:00 |
|
psy.c
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
psy.h
|
Removed the _t from all the celt*_t types to avoid clashing with POSIX
|
2009-10-17 22:12:42 -04:00 |
|
quant_bands.c
|
Doing the log2Amp() just in one place
|
2010-06-05 01:18:40 -04:00 |
|
quant_bands.h
|
Increasing log energy resolution
|
2010-03-13 10:04:19 -05:00 |
|
rangedec.c
|
Change ec_{enc|dec}_bit_prob to take probabilities in units of 1/65536 instead of 1/256. This allows them to use a single 16x16->32 multiply instead of a 24x8->32 multiply. Also change the time-frequency resolution flag coding to ensure that "0" is always the most-probable symbol (i.e., that prob("1")<50%), as that's where all the rounding error accumulates.
|
2010-05-29 23:45:28 -04:00 |
|
rangeenc.c
|
Change ec_{enc|dec}_bit_prob to take probabilities in units of 1/65536 instead of 1/256. This allows them to use a single 16x16->32 multiply instead of a 24x8->32 multiply. Also change the time-frequency resolution flag coding to ensure that "0" is always the most-probable symbol (i.e., that prob("1")<50%), as that's where all the rounding error accumulates.
|
2010-05-29 23:45:28 -04:00 |
|
rate.c
|
Allocation table now in bits/sample
|
2010-06-03 00:33:42 -04:00 |
|
rate.h
|
Changing the allocator resolution to 1/8 bit
|
2010-06-02 21:25:19 -04:00 |
|
stack_alloc.h
|
Updated copyright notices
|
2009-10-17 21:40:10 -04:00 |
|
testcelt.c
|
API change: optional resynthesis
|
2010-05-07 21:20:22 -04:00 |
|
vq.c
|
Splitting transients in time domain
|
2010-05-21 13:58:51 -04:00 |
|
vq.h
|
Cleaning up the folding code.
|
2010-05-09 19:45:31 -04:00 |
