shithub: opus

Info  •  Files  •  Log  •  Branches

33adba02 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/30 23:59

First version of pitch DNN C code

966a2d22 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/30 19:43

Code for 2D convolution

f3b86f94 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/29 22:48

Fix model saving

0459a572 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/29 11:34

updated PitchDNN export script

ce286958 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/29 11:31

refactoring and cleanup

49014454 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/29 10:34

fixed type in error message

c5c214df – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/29 10:25

added rudimentary support for dumping nn.Conv2d layers

25c65a0c – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/27 15:46

Fix stats indexing for state

9a7bb764 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/27 09:01

No features skip needed to align pitch features

a6b4fe37 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/27 09:00

Script to compute the groundtruth data using CREPE

217c40d4 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/27 08:57

dump cleanup, change alignment

ae0a140c – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/26 17:42

updated default values in adv_train_fargan.py

a71ba10f – Jean-Marc Valin <jmvalin@jmvalin.ca> authored and Jean-Marc Valin <jmvalin@amazon.com> committed on 2023/08/11 00:06

Fixes stack overflow for some custom modes

41a4c951 – Jan Buethe <jbuethe@amazon.de> authored and Jean-Marc Valin <jmvalin@amazon.com> committed on 2023/09/26 10:35

changed checkpoint format

733a095b – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/25 11:44

Adapting to new data format/model

f38b4a31 – Krishna Subramani <subramani.krishna97@gmail.com> authored and Jean-Marc Valin <jmvalin@amazon.com> committed on 2023/09/24 20:19

Python code for neural pitch

d88dd893 – Jean-Marc Valin <jmvalin@jmvalin.ca> authored and Jean-Marc Valin <jmvalin@amazon.com> committed on 2023/09/24 13:29

Add noise augmentation to pitch dumping

0100cd95 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/23 08:45

Add pitch feature computation

96d89e99 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/15 12:55

Band-limiting the voicing parameter

c4b83ae6 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/25 11:29

RDOVAE model update

574c766c – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/23 13:48

Infinite loops are bad

00580a63 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/22 07:39

bugfix

aad74fdf – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/22 07:18

pulse-related bugfix

0a92bc5e – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/21 11:01

more lavoce stuff

52c15629 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/15 20:59

Handle the case where the initial state didn't fit

71da9781 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/15 23:18

updated model

b88644b9 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/15 13:27

Quantizing initial state with rdovae too

2ec31cc5 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/20 15:06

added FARGAN adversarial training script

82f48d36 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/13 12:57

removed trailing whitespace in fargan

e7beaec3 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/13 12:31

integrated JM's FFT ada conv

b24c7b43 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/12 18:58

Remove --has-gain and --passthrough-size args

2f8b36d6 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/11 19:28

Add conditioning interpolation, fwconv layer

72c5ea41 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/06 13:15

Only use one frame of pre-loading

108b75c4 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/06 10:30

Randomly double the training sequence length

d54b9fb4 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/05 08:16

Adds skip connections

fb570ed8 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/02 20:52

5-tap pitch predictor

2e0c1ad3 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/09/01 09:14

Also use previous frame

4f63743f – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/31 12:32

explicit signal gain, explicit pitch predictor

1b13f631 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/30 14:36

FARGAN initial commit in Opus

4f4b6242 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/12 12:28

fix

5467539c – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/12 12:22

added requirements.txt to osce

2f290d32 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/12 10:50

added more enhancement stuff

7b8ba143 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/05 18:31

added copyright headers

35ee397e – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/05 08:29

added LPCNet torch implementation

90a171c1 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/05 08:13

brought NoLACE up to date

d431c321 – Michael Klingbeil <klingm@amazon.com> authored and Jean-Marc Valin <jmvalin@amazon.com> committed on 2023/09/01 19:18

Fixes vnni macro redefinition with clang

4a47b1a1 – Jan Buethe <jbuethe@amazon.de> authored on 2023/09/01 12:54

renamed ShapeNet to NoLACE

1fbdec87 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/09 14:28

Changing DRED exp. ID so we can reserve 127

bbe4dcc4 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/09 14:25

Don't allocate DRED bits if we don't have enough

0886828e – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/09 14:17

Making it easier to remove DRED experimental ID

5160d7fd – Jan Buethe <jbuethe@amazon.de> authored on 2023/08/08 06:46

improved auto-scaling in wexchange

6cba42f9 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/04 12:16

Add fwgan_load_model()

022f2b7e – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/04 10:34

int version for convenience

f7ee7137 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/03 21:32

Adding some comments

bd23d911 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/03 21:10

Continuation now working

6a184fc7 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/03 12:42

More scaling/filtering to run_fwgan_subframe()

b1a601fb – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/03 12:14

More FWGAN refactoring

5179896b – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/03 11:13

FWGAN refactoring

b0e1a2eb – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/02 22:06

Applying continuation after first subframe

3eac8c12 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/01 19:48

Avoid sin()/cos() calls for pitch reference

83e95a5c – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/08/01 17:26

Don't compute linear activation in-place

bf5eb5bf – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/30 22:06

Add FWGAN to lpcnet_demo

e62fd5c5 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/30 11:31

C implementation of FWGAN

155367d2 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/28 16:28

Skeleton for FWGAN code

e9f8402a – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/30 23:03

Handle float matrices with multiple of 8 rows

5eaa4a50 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/28 15:13

Add Gated Linear Unit (GLU)

5e045405 – Jan Buethe <jbuethe@amazon.de> authored on 2023/08/01 17:58

updated dump_model_weights.py

902d7636 – Jan Buethe <jbuethe@amazon.de> authored on 2023/08/01 14:18

added FWGAN weight dumping code

9691440a – Jan Buethe <jbuethe@amazon.de> authored on 2023/08/01 06:42

updadet wexchange version number

e916cf42 – Jan Buethe <jbuethe@amazon.de> authored on 2023/08/01 06:35

added .copy() to weights in wexchange

1fbc5fdd – Jan Buethe <jbuethe@amazon.de> authored on 2023/08/01 04:28

added auto-scaling to wexchange

aca390df – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/28 13:19

fixed wexchange for GRUs without bias

b50ddccf – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/28 14:29

Fixes weights parsing with DISABLE_DEBUG_FLOAT

8cc769ea – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/27 22:21

Switch RDO-VAE decoder to LinearLayer

ad057305 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/27 15:51

Make RDOVAE encoder use LinearLayer directly

eb72d29a – Jan Buethe <jbuethe@amazon.de> authored and Jean-Marc Valin <jmvalin@amazon.com> committed on 2023/07/24 13:13

Support for dumping LinearLayer in weight-exchange

b075eb53 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/25 23:16

oops, fix linear_init()

b1f94b1e – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/24 17:31

Add compute_generic_dense()

60d67b11 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/24 17:07

New compute_generic_conv1d()

4171532c – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/24 16:15

Add int8 type

101fd241 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/24 10:02

added dataset for SILK to LPCNet feature conversion

8e708090 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/23 14:25

Make float_weights optional

d15be43a – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/23 14:11

Make bias/subias/diag/scale optional

9d40e5cb – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/23 10:21

Add loading for LinearLayer

587c1020 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 11:16

clean-up

7487168d – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 10:51

added copyright headers

4f3761b0 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 10:47

added verbose option to run_test.py

0763a8f7 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 10:45

replaces multiprocessing module by multiprocess module in testsuite

81fe5f02 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 10:13

added matplotlib to testsuite requirements (needed by librosa)

f9aee675 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 09:31

added ShapeNet and ShapeUp48 models

57ab4949 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 09:19

removed whitespace

ba44bac4 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 09:10

added testsuite

0e5c103d – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 09:00

added weight-exchange library

8f7c72a6 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/21 17:42

Always define USE_SU_BIAS in vec_avx.h

4710bdf7 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/21 17:32

Add SSE2 support

9261eb5c – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/21 16:10

Refactoring to make VNNI and SSE2 easier

be5f58d6 – Jan Buethe <jbuethe@amazon.de> authored on 2023/07/22 07:11

neon related alignment requirement warning fix

cfc118d5 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/19 21:42

Merge branch 'dred_bitrate4' into opus-ng

62cd1c96 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/19 20:39

Transition to LinearLayer and remove unused code

f5a68a41 – Jean-Marc Valin <jmvalin@amazon.com> authored on 2023/07/15 22:21

Add generic linear layer