Skip to content

GitLab

Switch branch/tag
  1. 31 May, 2023 1 commit
  3. 26 May, 2023 2 commits
    • atalman's avatar
      Revert "Upgrade to FFmpeg5 (#3298)" (#3377) · 37779ef9
      atalman authored 
      Summary:
This reverts commit d38a7854.

This is temporary revert to unblock unit test migration from circleci to github

Pull Request resolved: https://github.com/pytorch/audio/pull/3377

Reviewed By: mthrok

Differential Revision: D46230498

Pulled By: atalman

fbshipit-source-id: 000d8a9ca00750fc1ca61f4c2cdd6e930a5ce46d
      37779ef9
    • Lakshmi Krishnan's avatar
      Improve RNN-T streaming decoding (#3295) · 9fc0dcaa
      Lakshmi Krishnan authored 
      Summary:
This commit fixes the following issues affecting streaming decoding quality
1. The `init_b` hypothesis is only regenerated from blank token if no initial hypotheses are provided.
2. Allows the decoder to receive top-K hypothesis to continue decoding from, instead of using just the top hypothesis at each decoding step.  This dramatically affects decoding quality especially for speech with long pauses and disfluencies.
3. Some minor errors regarding shape checking for length.

This also means that the resulting output is the entire transcript up until that time step, instead of just the incremental change in transcript.

Pull Request resolved: https://github.com/pytorch/audio/pull/3295

Reviewed By: nateanl

Differential Revision: D46216113

Pulled By: hwangjeff

fbshipit-source-id: 8f7efae28dcca4a052f434ca55a2795c9e5ec0b0
      9fc0dcaa
  5. 25 May, 2023 1 commit
    • Pingchuan Ma's avatar
      Add LRS3 AV-ASR recipe (#3278) · c6624fa6
      Pingchuan Ma authored 
      Summary:
This PR adds AV-ASR recipe which contains sample implementations of training and evaluation pipelines for RNNT based automatic, visual, and audio-visual (ASR, VSR, AV-ASR) models on LRS3. This repository includes both streaming/non-streaming modes.

CC stavros99 xiaohui-zhang YumengTao mthrok nateanl hwangjeff

Pull Request resolved: https://github.com/pytorch/audio/pull/3278

Reviewed By: nateanl

Differential Revision: D46121550

Pulled By: mpc001

fbshipit-source-id: bb44b97ae25e87df2a73a707008be46af4ad0fc6
      c6624fa6
  7. 23 May, 2023 1 commit
  9. 21 May, 2023 2 commits
  11. 16 May, 2023 1 commit
  13. 10 May, 2023 2 commits
  15. 05 May, 2023 1 commit
    • Zhaoheng Ni's avatar
      Update squim tutorial (#3313) · 05ef7dc6
      Zhaoheng Ni authored 
      Summary:
Add scatter plots for STOI, PESQ, Si-SDR, and MOS scores to demonstrate the performance of `SquimObjective` and `SquimSubjective` models and how close they are to the ground truths.

Pull Request resolved: https://github.com/pytorch/audio/pull/3313

Reviewed By: hwangjeff

Differential Revision: D45620311

Pulled By: nateanl

fbshipit-source-id: cb58ffd3744df4749b9385876da8de0cffd93557
      05ef7dc6
  17. 29 Apr, 2023 1 commit
  19. 28 Apr, 2023 1 commit
    • Yuekai Zhang's avatar
      Add cuctc decoder (#3096) · 0a1801ed
      Yuekai Zhang authored 
      Summary:
This PR implements a CUDA based ctc prefix beam search decoder.

Attach serveral benchmark results using V100 below:
|decoder type| model |datasets       | decoding time (secs)| beam size | batch size | model unit | subsampling times | vocab size |
|--------------|---------|------|-----------------|------------|-------------|------------|-----------------------|------------|
| cuctc |  conformer nemo    |dev clean        |7.68s | 8           |  32       | bpe         |    4  | 1000|
| cuctc |  conformer nemo   |dev clean  (sort by length)      |1.6s | 8           |  32       | bpe         |    4  | 1000|
| cuctc |  wav2vec2.0 torchaudio |dev clean                                |22s | 10           |  1       | char         |    2  | 29|
| cuctc |   conformer espnet   |aishell1 test                             | 5s | 10           |  24       | char         |    4  | 4233|

Note:
1.  The design is to parallel computation through batch and vocab axis, for loop the frames axis. So it's more friendly with smaller sequence lengths, larger vocab size comparing with CPU implementations.
2. WER is the same as CPU implementations. However, it can't decode with LM now.

Resolves: https://github.com/pytorch/audio/issues/2957.

Pull Request resolved: https://github.com/pytorch/audio/pull/3096

Reviewed By: nateanl

Differential Revision: D44709397

Pulled By: mthrok

fbshipit-source-id: 3078c54a2b44dc00eb4a81b4c657487eeff8c155
      0a1801ed
  21. 18 Apr, 2023 1 commit
  23. 31 Mar, 2023 1 commit
  25. 29 Mar, 2023 1 commit
    • moto's avatar
      Remove the note about AAC (#3214) · c07a96ab
      moto authored 
      Summary:
There is a part of StreamWriter tutorial that warns about corrupted AAC audio output, but this is no longer relevant thus this commit deletes it.

Pull Request resolved: https://github.com/pytorch/audio/pull/3214

Reviewed By: nateanl

Differential Revision: D44504030

Pulled By: mthrok

fbshipit-source-id: 4d26d582e9fb87d4e6fa674c05fe3192bc223eef
      c07a96ab
  27. 28 Mar, 2023 1 commit
  29. 16 Mar, 2023 1 commit
  31. 07 Mar, 2023 1 commit
    • Maciej Torhan's avatar
      Fix Adam and AdamW initializers in wav2letter example (#3145) · cea12eaf
      Maciej Torhan authored 
      Summary:
In wav2letter example there is passed `momentum` to `Adam` and `AdamW` initializer, which is not a correct parameter. To fix that we need to add `beta_1` and `beta_2` to arguments and replace `momentum` with them. I also added `eps` similar to `Adadelta` initializer.

Pull Request resolved: https://github.com/pytorch/audio/pull/3145

Reviewed By: mthrok

Differential Revision: D43847713

Pulled By: nateanl

fbshipit-source-id: 94f7c48232fabf520cfce81471694cb545d160c6
      cea12eaf
  33. 02 Mar, 2023 1 commit
  35. 24 Feb, 2023 2 commits
  37. 23 Feb, 2023 1 commit
    • G. Sun's avatar
      Add TCPGen context-biasing Conformer RNN-T (#2890) · 1ed330b5
      G. Sun authored 
      Summary:
This commit adds the implementation of the tree-constrained pointer generator (TCPGen) for contextual biasing.

An example for Librispeech can be found in audio/examples/asr/librispeech_biasing.

Maintainer's note (mthrok):
It seems that TrieNode should be better typed as tuple, but changing the implementation from list to tuple
could cause some issue without running the code, so the code is not changed, though the annotation uses tuple.

Pull Request resolved: https://github.com/pytorch/audio/pull/2890

Reviewed By: nateanl

Differential Revision: D43171447

Pulled By: mthrok

fbshipit-source-id: 372bb077d997d720401dbf2dbfa131e6a958e37e
      1ed330b5
  39. 16 Feb, 2023 2 commits
  41. 15 Feb, 2023 1 commit
  43. 14 Feb, 2023 1 commit
    • Zhaoheng Ni's avatar
      Update ssl example (#3060) · ff01be0f
      Zhaoheng Ni authored 
      Summary:
- Rename the current `ssl` example to `self_supervised_learning`
- Add README to demonstrate how to run the recipe with hubert task

Pull Request resolved: https://github.com/pytorch/audio/pull/3060

Reviewed By: mthrok

Differential Revision: D43287868

Pulled By: nateanl

fbshipit-source-id: 10352682485ef147ca32f4c4c9f9cde995444aa0
      ff01be0f
  45. 30 Jan, 2023 1 commit
    • Yan Li's avatar
      Fix hybrid demucs tutorial for CUDA (#3017) · da9d1627
      Yan Li authored 
      Summary:
Currently there will be a few errors when this tutorial is run with a CUDA device.

The reasons being:
- The source audio waveform is not properly moved to the GPU. The `to()` method is not in-place for Tensors, so we need to assign the return value of the method call to the variable (otherwise the Tensor would still be on the CPU).
- When performing further analysis and displaying of the output audio, we need to move them back from the GPU to the CPU. This is because some of the functions we call require the Tensor to be on the CPU (e.g. `stft()` and `bss_eval_sources()`).

Pull Request resolved: https://github.com/pytorch/audio/pull/3017

Reviewed By: mthrok

Differential Revision: D42828526

Pulled By: nateanl

fbshipit-source-id: c28bc855e79e3363a011f4a35a69aae1764e7762
      da9d1627
  47. 19 Jan, 2023 2 commits
    • Zhaoheng Ni's avatar
      Add modularized SSL training recipe (#2876) · 2eaefe27
      Zhaoheng Ni authored 
      Summary:
TorchAudio currently has one training recipe for HuBET + LibriSpeech pre-training. It may not suit well when users want to use customized dataset, or use a new training objective (such as contrastive loss in Wav2Vec2). The PR addresses the issue by providing a modularized training recipe for audio self-supervised learning. Users can inject customized model module, loss function, optimizer, lr scheduler, and datamodule for training a SSL model.

Pull Request resolved: https://github.com/pytorch/audio/pull/2876

Reviewed By: hwangjeff

Differential Revision: D42617414

Pulled By: nateanl

fbshipit-source-id: 6413df45a9d106ed1d5ff830bf628c54368c5792
      2eaefe27
    • hwangjeff's avatar
      Simplify train step in Conformer RNN-T LibriSpeech recipe (#2981) · c6a52355
      hwangjeff authored 
      Summary:
In the Conformer RNN-T LibriSpeech recipe, there's no need to perform manual optimization. This PR modifies the recipe to use automatic optimization instead.

Pull Request resolved: https://github.com/pytorch/audio/pull/2981

Reviewed By: mthrok

Differential Revision: D42507228

Pulled By: hwangjeff

fbshipit-source-id: 9712add951eba356e39f7e8c8dc3bf584ba48309
      c6a52355
  49. 17 Jan, 2023 1 commit
  51. 16 Jan, 2023 1 commit
    • Robin Scheibler's avatar
      Fixes examples/source_separation for WSJ0_2mix dataset (#2987) · f9d38796
      Robin Scheibler authored 
      Summary:
The `examples/source_separation` scripts use inconsistent keyword to indicate the WSJ0_2mix dataset. This PR does the following.

1. Use `wsj0mix` consistently as keyword indicating the WSJ0_2mix dataset
2. Corrects `args.data_dir` to `args.root_dir` in eval.py
3. Modify the parameters of `pytorch_lightning.Trainer` according to latest version (use `accelerator="gpu"` and `devices=args.num_devices`, instead of just `gpus=args.num_devices`)

Pull Request resolved: https://github.com/pytorch/audio/pull/2987

Reviewed By: xiaohui-zhang

Differential Revision: D42536992

Pulled By: nateanl

fbshipit-source-id: 10a80263ad7054b1629d8fa023676b607e633d76
      f9d38796
  53. 13 Jan, 2023 1 commit
  55. 30 Dec, 2022 1 commit
  57. 17 Dec, 2022 1 commit
  59. 16 Dec, 2022 1 commit
    • Caroline Chen's avatar
      Rename resampling_method options (#2922) · e6bebe6a
      Caroline Chen authored 
      Summary:
resolves https://github.com/pytorch/audio/issues/2891

Rename `resampling_method` options to more accurately describe what is happening. Previously the methods were set to `sinc_interpolation` and `kaiser_window`, which can be confusing as both options actually use sinc interpolation methodology, but differ in the window function used. As a result, rename `sinc_interpolation` to `sinc_interp_hann` and `kaiser_window` to `sinc_interp_kaiser`. Using an old option will throw a warning, and those options will be deprecated in 2 released. The numerical behavior is unchanged.

Pull Request resolved: https://github.com/pytorch/audio/pull/2922

Reviewed By: mthrok

Differential Revision: D42083619

Pulled By: carolineechen

fbshipit-source-id: 9a9a7ea2d2daeadc02d53dddfd26afe249459e70
      e6bebe6a
  61. 29 Nov, 2022 1 commit
  63. 28 Nov, 2022 1 commit
  65. 17 Nov, 2022 1 commit
  67. 16 Nov, 2022 1 commit