shithub: opus

c5e04e4a – Jean-Marc Valin <[email protected]> authored on 2013/03/12 07:24

Makes bandwidth detection less aggressive.

f982b84d – Jean-Marc Valin <[email protected]> authored on 2013/03/11 12:25

Fixes a minor glitch on SILK bandwidth changes

42f39c54 – Ralph Giles <[email protected]> authored on 2013/03/11 06:59

Add -Wdeclaration-after-statement.

933a2754 – Koen Vos <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2013/03/11 09:30

High-band attenuation tuning for hybrid mode

5a6e7c8b – Jean-Marc Valin <[email protected]> authored on 2013/03/11 09:41

Fixes pseudo-stack

22345ef1 – Taihei Momma <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2013/03/11 09:07

Fixes OPUS_SET_EXPERT_FRAME_DURATION_REQUEST for multi-stream

097fd4b1 – Robert Meakins <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2013/03/11 08:59

Fixes C90 issues for obsolete compilers

d865fe61 – Jean-Marc Valin <[email protected]> authored on 2013/03/11 07:35

Adds stereo width computation

f96fc8cc – Jean-Marc Valin <[email protected]> authored on 2013/03/08 07:29

Merge branch 'exp_analysis'

fa43c770 – Jean-Marc Valin <[email protected]> authored on 2013/03/08 06:35

Making CELT stereo<->mono transitions gradual

f55b0eac – Jean-Marc Valin <[email protected]> authored on 2013/03/07 13:13

Documentation fixes

a41a585f – James Zern <[email protected]> authored and Gregory Maxwell <[email protected]> committed on 2013/02/27 08:16

test_opus_decode: force integer constants unsigned

d9aa6e04 – Jean-Marc Valin <[email protected]> authored on 2013/03/01 11:07

Oops, fixed API name from previous commit

32c4a0c9 – Jean-Marc Valin <[email protected]> authored on 2013/03/01 10:18

Applies soft-clipping to the int decoder API.

73142b10 – Jean-Marc Valin <[email protected]> authored on 2013/02/28 10:30

Makes the speech/music probability estimation mode conservative

742aac10 – Jean-Marc Valin <[email protected]> authored on 2013/02/22 11:44

Adds silence probability to speech/music detector

74f36b56 – Jean-Marc Valin <[email protected]> authored on 2013/02/20 17:31

oops s/IMAX/IMIN/

51f4a32e – Jean-Marc Valin <[email protected]> authored on 2013/02/19 23:08

Adds support for delayed decision

69c3dcd1 – Jean-Marc Valin <[email protected]> authored on 2013/02/18 22:42

Fixes a bug introduced in 851f803

3ecd9c20 – Jean-Marc Valin <[email protected]> authored on 2013/02/18 19:42

Fixes two bandwidth decision issues

851f8033 – Jean-Marc Valin <[email protected]> authored on 2013/02/17 20:43

Moves mono downmixing and upsampling to the compute_mdcts()

541a472f – Jean-Marc Valin <[email protected]> authored on 2013/02/17 16:21

Adds code to revise the transient decision using the band energy

10a34a5d – Jean-Marc Valin <[email protected]> authored on 2012/12/19 19:23

Making multistream variable duration work for both the float and int API

bb43b8b6 – Jean-Marc Valin <[email protected]> authored on 2012/12/17 13:02

No need for extra_buffer anymore

95561be6 – Jean-Marc Valin <[email protected]> authored on 2012/12/17 12:54

Better handling of the multistream bitrate

74483660 – Jean-Marc Valin <[email protected]> authored on 2012/12/17 11:23

Multistream support for variable frame duration

f548a5a3 – Jean-Marc Valin <[email protected]> authored on 2012/11/26 18:20

Makes variable framesize less aggressive at lower rates

85479648 – Jean-Marc Valin <[email protected]> authored on 2012/11/25 22:08

Re-enable analysis for 40- and 60-ms frames

49583ed6 – Jean-Marc Valin <[email protected]> authored on 2012/11/22 08:11

First attempt at varying the frame size depending on the audio (float only)

e85a6f5c – Jean-Marc Valin <[email protected]> authored on 2012/11/19 18:21

Makes opus_demo rubust to the encoder using variable frame duration

2a5f0565 – Jean-Marc Valin <[email protected]> authored on 2012/11/19 18:17

Running transient_analysis() even for 2.5 ms frames

48ac1221 – Jean-Marc Valin <[email protected]> authored on 2012/11/13 21:39

Makes analysis usable for all frame sizes

7ebacf43 – Jean-Marc Valin <[email protected]> authored on 2012/11/12 21:24

Moves analysis to the beginning of opus_encode()

6044907c – Jean-Marc Valin <[email protected]> authored on 2013/02/07 10:31

Fixes a SILK fixed-point encoder issue reported by Cliff Parris <[email protected]>

e51a3f33 – Jean-Marc Valin <[email protected]> authored on 2013/02/06 18:48

Fixes a minor CELT->SILK switching glitch in the decoder

61e9eb15 – Jean-Marc Valin <[email protected]> authored on 2013/02/06 13:10

Adds a ramp up in the SILK prefill

4ed7e48e – Jean-Marc Valin <[email protected]> authored on 2013/02/04 20:32

Fixes a hybrid stereo encoder issue

2fd3d0aa – Ralph Giles <[email protected]> authored on 2013/01/16 11:30

Split mapping families into sections and add dowmix matrices.

0431f933 – Ralph Giles <[email protected]> authored on 2013/01/16 09:14

Improve Content Type figure placement.

e7aa6cb9 – Ralph Giles <[email protected]> authored on 2013/01/11 12:11

Minor clarifications to the channel mapping section.

fc113b73 – Ralph Giles <[email protected]> authored on 2013/01/11 12:01

Use 'front center' in the channel mapping table.

17d7fed1 – Ralph Giles <[email protected]> authored on 2013/01/11 10:51

Fix a typo.

4fec6465 – Timothy B. Terriberry <[email protected]> authored on 2013/01/10 14:29

Add HTTPS to make gmaxwell happy.

2d477a93 – Timothy B. Terriberry <[email protected]> authored on 2013/01/10 14:23

Flesh out the Implementation Status section a bit.

a1b913f5 – Ralph Giles <[email protected]> authored on 2013/01/10 12:00

Clarify how grouped streams are not "Ogg Opus files".

0c1487ae – Ralph Giles <[email protected]> authored on 2013/01/10 11:38

Add an 'Implementation Status' section to the Ogg Opus draft.

7ad2f439 – Ralph Giles <[email protected]> authored on 2013/01/10 09:15

Trim an irrelevant clause.

8e183a32 – Jean-Marc Valin <[email protected]> authored on 2013/01/10 04:44

RTP draft adopted as WG item

ebdfbfb7 – Jean-Marc Valin <[email protected]> authored on 2013/01/09 06:13

Fixes stupid preprocessor-related bug introduced in e368e6209.

fa28dea7 – Jean-Marc Valin <[email protected]> authored on 2013/01/03 16:15

Fixes a seed issue introduced in 5367dac

e368e620 – Jean-Marc Valin <[email protected]> authored on 2013/01/03 09:28

Remove condition in pre/de-emphasis when not building with custom modes

ba1bd031 – Ralph Giles <[email protected]> authored on 2013/01/02 10:28

Use AC_CONFIG_HEADERS.

5367dac3 – Jean-Marc Valin <[email protected]> authored on 2012/12/22 16:56

seed and remaining_bits moved to band context

a9d6286c – Jean-Marc Valin <[email protected]> authored on 2012/12/22 16:32

Using a band context to reduce the number of arguments being passed around.

f18acee4 – Jean-Marc Valin <[email protected]> authored on 2012/12/19 07:39

more bands.c cleaning up

ddd0d214 – Jean-Marc Valin <[email protected]> authored on 2012/12/19 06:37

collapse mask cleanup

414fd270 – Jean-Marc Valin <[email protected]> authored on 2012/12/19 06:25

Removes useless parameters and re-indents the code

3d6c3418 – Jean-Marc Valin <[email protected]> authored on 2012/12/19 06:04

Splitting off the recursion in quant_partition()

4ffbf211 – Jean-Marc Valin <[email protected]> authored on 2012/12/19 05:26

stereo part of quant_band() moved to quant_band_stereo()

113a5e1b – Jean-Marc Valin <[email protected]> authored on 2012/12/19 04:58

Moves the split angle code to compute_theta()

dfe3bf96 – Jean-Marc Valin <[email protected]> authored on 2012/12/19 02:19

First step in quant_band() cleanup: N=1 case.

d683c76b – Jean-Marc Valin <[email protected]> authored on 2012/12/21 11:17

Fixes MSVC warnings for double->float and float->int conversion

b5f73945 – Ralph Giles <[email protected]> authored on 2012/12/21 07:06

Add float literal suffixes to mlp_data.c.

724385e2 – Ralph Giles <[email protected]> authored on 2012/12/21 06:56

Add float suffix to tansig_table literals.

0a29bb51 – Ralph Giles <[email protected]> authored on 2012/12/21 06:49

Add tansig_table.h to the MSVC project files.

3593bbf4 – Jean-Marc Valin <[email protected]> authored on 2012/12/21 09:11

missing comment in win32/config.h

23fcd706 – Jean-Marc Valin <[email protected]> authored on 2012/12/21 08:09

Only use automatic bandwidth detection when the user doesn't force bandwidth

2dd3fb9d – Jean-Marc Valin <[email protected]> authored on 2012/12/21 07:28

Fixes bandwidth detection by not relying on uninitialized data

5fb50adb – Jean-Marc Valin <[email protected]> authored on 2012/12/20 20:23

skip bands with no energy

b3eba24b – Jean-Marc Valin <[email protected]> authored on 2012/12/20 18:11

Capping lsb_depth to 16 unless using the float API of a float build

7509fdb8 – Jean-Marc Valin <[email protected]> authored on 2012/12/20 17:48

New bandwidth detection code

56921ff7 – Jean-Marc Valin <[email protected]> authored on 2012/12/20 08:00

Bump version number

05bf400a – Ralph Giles <[email protected]> authored on 2012/12/19 07:21

Update ogg draft based on comments from Tina le Grand.

c763a35e – Ralph Giles <[email protected]> authored on 2012/12/18 11:37

Update the oggopus script to use xml2rfc v2.

90bac9d1 – Jean-Marc Valin <[email protected]> authored on 2012/12/14 09:01

Minor tuining of complexity thresholds

28ef277d – Jean-Marc Valin <[email protected]> authored on 2012/12/14 08:21

VBR code simplification

dae16fb9 – Jean-Marc Valin <[email protected]> authored on 2012/12/13 16:40

Cleanup: tf_estimate now ranges from 0 to 1

f8809dda – Jean-Marc Valin <[email protected]> authored on 2012/12/12 10:22

Dynalloc rate calibration

1122d299 – Jean-Marc Valin <[email protected]> authored on 2012/12/12 10:16

Makes "stereo savings" less aggressive on mono or near-mono.

b33db8f3 – Jean-Marc Valin <[email protected]> authored on 2012/12/12 09:45

Makes tonality boost less aggressive

41fd7a12 – Jean-Marc Valin <[email protected]> authored on 2012/12/12 09:41

Disables the new spread+tapset estimator

256c9595 – Ralph Giles <[email protected]> authored on 2012/12/11 05:17

Document that opus_packet_get_nb_frames, etc. can return OPUS_BAD_ARG.

8eb9bb79 – Timothy B. Terriberry <[email protected]> authored on 2012/12/07 09:07

Numerous PLC cleanups.

22d8dcf3 – Ralph Giles <[email protected]> authored on 2012/12/07 04:16

Remove the obsolete CELT_BUILD define from win32/config.h.

19239ea3 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2012/12/05 16:43

Add tests for GET_LAST_PACKET_DURATION.

a0737d1f – Jean-Marc Valin <[email protected]> authored on 2012/12/05 16:48

Fixes OPUS_GET_LAST_PACKET_DURATION

535efb07 – Gregory Maxwell <[email protected]> authored on 2012/12/05 14:26

Add missing packet_get_nb_samples test coverage and fix test output.

119d53c9 – Ralph Giles <[email protected]> authored on 2012/12/05 10:57

Bump soname version to match the 1.0.x branch.

d215a113 – Ralph Giles <[email protected]> authored on 2012/12/05 06:57

Link to the new Microsoft patent license.

2c1a11ff – Jean-Marc Valin <[email protected]> authored on 2012/12/04 19:54

Adds assert to catch bug from previous commit

8c9c9b28 – Jean-Marc Valin <[email protected]> authored on 2012/12/04 19:53

Fixes a multi-frame FEC issue that was caught by valgrind

9283114f – Jean-Marc Valin <[email protected]> authored on 2012/12/04 19:50

Adds explicit valgrind checks when ENABLE_VALGRIND is defined

d0fd9d4b – Jean-Marc Valin <[email protected]> authored on 2012/12/04 10:45

Implements opus_packet_get_nb_samples()

512d849c – Jean-Marc Valin <[email protected]> authored on 2012/12/04 09:13

Implements OPUS_GET_LAST_FRAME_DURATION decoder ctl()

a5bd4409 – Jean-Marc Valin <[email protected]> authored on 2012/12/04 09:13

Don't update the internal decoder state until we know the packet is valid

7fcd66c4 – Jean-Marc Valin <[email protected]> authored on 2012/12/04 10:07

Changes the PLC behaviour and fixes the FEC behaviour on concatenated packets

124f69b0 – Ralph Giles <[email protected]> authored on 2012/12/03 08:46

Remove MSVC incremental linking directives.

c093829c – Nils Wallménius <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2012/12/02 11:06

Do imdct post-rotate and deshuffle in-place.

aa3a599c – Jean-Marc Valin <[email protected]> authored on 2012/12/03 10:40

Fixes forward MDCT for overlaps that aren't a multiple of 4

259e1666 – Jean-Marc Valin <[email protected]> authored on 2012/12/03 08:05

Fix packet length handling for 16-bit machines (makes no difference on 32-bit)