shithub: opus

Info  •  Files  •  Log  •  Branches

cfc8e300 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2017/02/13 20:32

Rename y0/y1 to out0/out1 to avoid name clash with the Bessel functions

468327d7 – Felicia Lim <flim@google.com> authored on 2017/02/10 06:42

Add decoder fuzz target.

572d65df – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2017/02/09 18:01

Remove silk_LPC_inverse_pred_gain_Q24() which is no longer used anywhere

c5a84df8 – Mark Harris <mark.hsj@gmail.com> authored on 2017/02/04 11:57

arm2gnu.pl: Fix GNU ARM .type directive

feabcdc6 – James Zern <jzern@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2017/02/03 13:55

fix ARM build w/--disable-intrinsics --enable-asm

cfdaf365 – Linfeng Zhang <linfengz@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/08/25 14:12

Optimize silk_NSQ_del_dec() for ARM NEON

68afa490 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2017/01/25 16:56

Avoids undefined behaviour in ARM-optimized code

32b302c4 – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2017/01/18 09:47

Add --enable-check-asm

864bc8c6 – Linfeng Zhang <linfengz@google.com> authored and Felicia Lim <flim@google.com> committed on 2016/08/11 09:21

Update NSQ_LPC_BUF_LENGTH macro

aef47564 – Mark Harris <mark.hsj@gmail.com> authored on 2017/01/17 19:13

configure: Improve error message

d6643192 – Felicia Lim <flim@google.com> authored on 2017/01/17 11:23

Add FIXED_POINT check if using celt_fir in SILK's LPC analysis filter

783ad767 – Linfeng Zhang <linfengz@google.com> authored and Felicia Lim <flim@google.com> committed on 2016/06/16 12:21

Revise celt_fir_c() to not pass in argument "mem"

c9ba5520 – James Zern <jzern@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2017/01/12 15:29

fixed_debug.h: delete some dead code

19b13443 – Chris Rudmin <chris.rudmin@gmail.com> authored and Timothy B. Terriberry <tterribe@xiph.org> committed on 2017/01/03 12:03

Fix error message

ec11c822 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2017/01/05 18:39

Avoids pre-echo in hybrid mode caused by noise being injected in the first band

efcdeb9b – Mark Harris <mark.hsj@gmail.com> authored on 2016/12/22 06:11

Fix some compiler warnings

cf9409fe – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/12/16 12:52

Makes analysis run at 24 kHz, with 20-ms frames

159bb6df – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/12/19 11:15

update draft: oops, fixes use of "above"

4a48ce89 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/12/19 09:33

Update draft: addressing WGLC comments

c84f4866 – Mark Harris <mark.hsj@gmail.com> authored on 2016/12/17 13:32

opus_demo: remove obsolete option from usage

83758306 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/12/16 09:11

Adding a missing "else"

8fe210f1 – Ricardo Constantino <wiiaboo@gmail.com> authored and Ralph Giles <giles@thaumas.net> committed on 2016/11/04 22:03

genversion.bat: generate same version as update_version

bd54e279 – Mark Harris <mark.hsj@gmail.com> authored on 2016/11/16 15:03

Correct SILK encoder gain limit

ab148485 – Ricardo Constantino <wiiaboo@gmail.com> authored and Ralph Giles <giles@thaumas.net> committed on 2016/11/03 18:26

vs2015: reenable DebugDLL versions of opus_demo

4dde082d – Ralph Giles <giles@thaumas.net> authored on 2016/11/03 11:40

Remove SILK_DEBUG define when _WIN32 and _DEBUG are defined.

3fdfcd40 – Ralph Giles <giles@thaumas.net> authored on 2016/11/03 11:25

Remove commented-out DEBUG_STORE_DATA calls. r=keon

f4389ef7 – Ricardo Constantino <wiiaboo@gmail.com> authored and Ralph Giles <giles@thaumas.net> committed on 2016/11/10 16:27

appveyor: don't leave artifacts under a subdir

71ae28ab – Ricardo Constantino <wiiaboo@gmail.com> authored and Ralph Giles <giles@thaumas.net> committed on 2016/11/10 16:27

appveyor: pack interesting files too, not just lib

d73c5a96 – Mark Harris <mark.hsj@gmail.com> authored on 2016/11/05 18:01

Remove redundant code

d74fa278 – Mark Harris <mark.hsj@gmail.com> authored on 2016/11/05 17:32

Fix crash on bad encoder frame_size argument

132ed594 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/11/04 07:41

Removes OPUS_FRAMESIZE_VARIABLE

e619ea76 – Ricardo Constantino <wiiaboo@gmail.com> authored and Ralph Giles <giles@thaumas.net> committed on 2016/11/03 17:12

vs2015: Add missing file to test_opus_encode.

bd109a2b – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/11/03 10:39

Adds checksum for 1.2-alpha

92282c1a – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/11/02 13:59

Improves CELT bit allocation at very low bitrates

f1deb29c – Mark Harris <mark.hsj@gmail.com> authored on 2016/11/02 02:56

Eliminate trailing spaces

20bf3b33 – Mark Harris <mark.hsj@gmail.com> authored on 2016/11/02 02:18

Reduce redundancy when SILK uses too many bits

1fd53f9a – Mark Harris <mark.hsj@gmail.com> authored on 2016/11/02 02:16

Fix redundancy on SILK 80+ms bandwidth switch

d31f883c – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/11/01 15:00

Add copyright notice to autogen.sh to make the Chromium people happy

80052fe3 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/11/01 14:41

Fixes NE10 configure problem

b0949f11 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/11/01 12:10

Temporarily disable -Wnonnull around an opus_decode() call in test_opus_decode.c

1b633ab0 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/31 17:49

Bump static PACKAGE_VERSION string in version.mk

8fec416b – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/10/31 07:58

Fix amount of extra bytes reserved for 100 ms

7e122394 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/29 13:02

Don't use hybrid "weak transients" on vowels

e8b6ee2d – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/29 12:46

Adds opus_encode_regressions.c to the MIPS Makefile

880f57a4 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/29 12:10

Adds opus_encode_regressions.c to the UNIX Makefile

d0d1cf3b – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/29 10:45

New regression test from Mark's fuzzing test cases

90f20c62 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/28 12:44

Define "weak transients" for low bitrate hybrid

bcfe0f68 – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/10/28 07:11

Multistream encoder: allocate an extra byte per stream for 100 ms

e564cc50 – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/10/28 05:34

Fix CBR packet sizes for 80 ms

33698ee1 – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/10/26 12:48

Update bandwidth and opusCanSwitch before generating SILK DTX ToC

9962bf3d – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/10/27 13:00

Encode correct headers when asking for very small output bytes and using long frames

0962cbe2 – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/10/27 13:03

Support encoding 80/100/120 ms frame lengths

8bca154b – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/09/22 07:06

Move multiframe encoding into a separate function

2af92cd9 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/27 10:14

Fixes an unstable energy issue for low-bitrate hybrid

76fa939b – Ralph Giles <giles@thaumas.net> authored on 2016/10/27 06:23

Make use of dot optional in generating documentation.

6d062849 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/26 20:11

Account for redundancy signalling when computing st->silk_mode.maxBits

4667b699 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/20 17:56

updating update draft

38688317 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/10/04 18:07

Fixing some opus_int vs opus_int32 mismatches

ae5f5cc1 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/19 21:44

Using OPUS_MOVE() instead of OPUS_COPY() to move redundancy bytes

8ac5d899 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/19 13:18

Consider per-channel energy bits in surround_rate_allocation()

a3fadcb9 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/09 12:13

Fixes surround_analysis() for frame size > 20 ms

c305b824 – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/09/16 07:12

Remove float ops from encoder settings fuzzer test

f4dcc09f – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/15 11:02

Fix mixed declarations and code

348bb27a – Felicia Lim <flim@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/09/15 07:23

Fuzzer test for changing encoder settings online

e5ad2616 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/14 07:38

Prevents ridiculously large gains from causing inf/NaNs in float decoder

70e3c348 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/09 12:51

Fixing failure due to CBR allocating zero byte to a stream

fa2578bf – Michael Graczyk <mgraczyk@google.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/09/07 11:26

Fix ambisonics bitrate when bitrate_bps is OPUS_AUTO

ac9fabc7 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/07 18:04

Fixes an internal error in the Opus encoder

dbff5fca – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/07 07:20

Prevent tonality estimator from running too often with 2.5 and 5 ms frames

f4f5a6a0 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/06 18:32

Prevents an overflow in multi-stream encoder rate computation

691d4935 – Mark Harris <mark.hsj@gmail.com> authored on 2016/09/03 06:42

Allow disabling phase inversion for multistream

f25b46a3 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/01 22:36

misc fixes to update draft

cf0e557a – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/01 10:33

minor fix to update draft

08eb1b8b – Ricardo Constantino (:RiCON) <wiiaboo@gmail.com> authored and Jean-Marc Valin <jmvalin@jmvalin.ca> committed on 2016/09/01 15:21

vs2015: include files added in 76674fea

f9a57554 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/09/01 04:34

Adds two integer wrap-around issues to the update draft

18a380a7 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/04/20 09:27

Make it possible to ignore inverted phase stereo for downmix purposes

6db0908d – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/06/06 09:44

Bitstream change: prevent noise injection in hybrid mode

12fb15f4 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/30 21:57

Add configure option --enable-update-draft

7d0e33f7 – Mark Harris <mark.hsj@gmail.com> authored on 2016/08/27 03:10

Correct multistream variable frame size stride

3421f6d2 – Mark Harris <mark.hsj@gmail.com> authored on 2016/08/27 03:03

Correct analysis downmix scaling factor

9f3883a9 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/25 10:24

Properly process both channels for DTX

096b2207 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/21 11:53

Make the DC rejection filter Q14 instead of Q15

8ab3e80e – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/07/30 18:40

Fixes an overflow in LPC_inverse_pred_gain_QA()

18335e8e – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/15 15:50

Removed a float operation that sneaked in in the fixed-point code

b4d0df8d – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/15 15:19

Oops, fixes the sign in dc_reject()

adf94f2c – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/15 14:24

Implementing compute_frame_energy() from celt_inner_prod() for float

b7c76539 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/15 13:30

Speeding up PVQ search by allocating even more pulses in the projection.

fb19ba91 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/15 13:26

Reducing the overhead due to dependency chains in dc_reject() for stereo

9ac94266 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/15 12:46

Reducing the dependency chain in dc_reject()

11dba8d9 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/14 13:56

Apply deemphasis to both channels in the same loop for the simple case

cd159fd1 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/12 22:53

Making gcc use SSE directly for float->int conversion when available

9f7e502e – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/12 19:49

Reducing dependencies in deemphasis()

7f3fb201 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/11 19:23

Fixes problem with M_PI

2ff6556f – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/11 07:05

Making stereo_itheta() use the same atan2() approximation as tonality_analysis()

76674fea – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/09 19:22

SSE2 implementation of the PVQ search

e806d6a7 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/06 11:49

Making signx[] an int in alg_quant() and removes unnecessary sign copying

c7bbc3e3 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/05 21:01

Speeding up PVQ using unlikely() and moving first position out of the loop

09b53529 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/05 20:06

Getting gcc to use cmovs rather than branches in alg_quant()

416611cf – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/07/29 09:47

biasing quantization

a01feba0 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored on 2016/08/05 13:34

Fixes "mixed declarations and code"