Skip to content

GitLab

Switch branch/tag
  1. 06 Jun, 2023 1 commit
  3. 26 May, 2023 1 commit
    • 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. 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
  9. 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
  11. 16 Feb, 2023 1 commit
  13. 19 Jan, 2023 1 commit
  15. 17 Nov, 2022 1 commit
  17. 09 Sep, 2022 1 commit
  19. 10 Aug, 2022 1 commit
  21. 11 Jul, 2022 1 commit
  23. 23 Jun, 2022 1 commit
  25. 04 Jun, 2022 1 commit
  27. 01 Jun, 2022 1 commit
    • Caroline Chen's avatar
      Move CTC beam search decoder to beta (#2410) · 93024ace
      Caroline Chen authored 
      Summary:
Move CTC beam search decoder out of prototype to new `torchaudio.models.decoder` module.

hwangjeff mthrok any thoughts on the new module + naming, and if we should move rnnt beam search here as well??

Pull Request resolved: https://github.com/pytorch/audio/pull/2410

Reviewed By: mthrok

Differential Revision: D36784521

Pulled By: carolineechen

fbshipit-source-id: a2ec52f86bba66e03327a9af0c5df8bbefcd67ed
      93024ace
  29. 15 May, 2022 1 commit
    • John Reese's avatar
      [codemod][usort] apply import merging for fbcode (8 of 11) · d62875cc
      John Reese authored 
      Summary:
Applies new import merging and sorting from µsort v1.0.

When merging imports, µsort will make a best-effort to move associated
comments to match merged elements, but there are known limitations due to
the diynamic nature of Python and developer tooling. These changes should
not produce any dangerous runtime changes, but may require touch-ups to
satisfy linters and other tooling.

Note that µsort uses case-insensitive, lexicographical sorting, which
results in a different ordering compared to isort. This provides a more
consistent sorting order, matching the case-insensitive order used when
sorting import statements by module name, and ensures that "frog", "FROG",
and "Frog" always sort next to each other.

For details on µsort's sorting and merging semantics, see the user guide:
https://usort.readthedocs.io/en/stable/guide.html#sorting

Reviewed By: lisroach

Differential Revision: D36402214

fbshipit-source-id: b641bfa9d46242188524d4ae2c44998922a62b4c
      d62875cc
  31. 12 May, 2022 1 commit
  33. 11 May, 2022 1 commit
    • hwangjeff's avatar
      Refactor LibriSpeech Conformer RNN-T recipe (#2366) · 69467ea5
      hwangjeff authored 
      Summary:
Modifies the example LibriSpeech Conformer RNN-T recipe as follows:
- Moves data loading and transforms logic from lightning module to data module (improves generalizability and reusability of lightning module and data module).
- Moves transforms logic from dataloader collator function to dataset (resolves dataloader multiprocessing issues on certain platforms).
- Replaces lambda functions with `partial` equivalents (resolves pickling issues in certain runtime environments).
- Modifies training script to allow for specifying path model checkpoint to restart training from.

Pull Request resolved: https://github.com/pytorch/audio/pull/2366

Reviewed By: mthrok

Differential Revision: D36305028

Pulled By: hwangjeff

fbshipit-source-id: 0b768da5d5909136c55418bf0a3c2ddd0c5683ba
      69467ea5
  35. 21 Apr, 2022 1 commit
    • hwangjeff's avatar
      Change underlying implementation of RNN-T hypothesis to tuple (#2339) · 6b242c29
      hwangjeff authored 
      Summary:
PyTorch Lite, which is becoming a standard for mobile PyTorch usage, does not support containers containing custom classes. Consequently, because TorchAudio's RNN-T decoder currently returns and accepts lists of `Hypothesis` namedtuples, it is not compatible with PyTorch Lite. This PR resolves said incompatibility by changing the underlying implementation of `Hypothesis` to tuple.

Pull Request resolved: https://github.com/pytorch/audio/pull/2339

Reviewed By: nateanl

Differential Revision: D35806529

Pulled By: hwangjeff

fbshipit-source-id: 9cbae5504722390511d35e7f9966af2519ccede5
      6b242c29
  37. 13 Apr, 2022 1 commit
    • hwangjeff's avatar
      Add Conformer RNN-T LibriSpeech training recipe (#2329) · c262758b
      hwangjeff authored 
      Summary:
Adds Conformer RNN-T LibriSpeech training recipe to examples directory.

Produces 30M-parameter model that achieves the following WER:

|                     |          WER |
|:-------------------:|-------------:|
| test-clean          |       0.0310 |
| test-other          |       0.0805 |
| dev-clean           |       0.0314 |
| dev-other           |       0.0827 |

Pull Request resolved: https://github.com/pytorch/audio/pull/2329

Reviewed By: xiaohui-zhang

Differential Revision: D35578727

Pulled By: hwangjeff

fbshipit-source-id: afa9146c5b647727b8605d104d928110a1d3976d
      c262758b
  39. 04 Apr, 2022 2 commits
  41. 17 Feb, 2022 1 commit
  43. 16 Feb, 2022 6 commits
  45. 11 Feb, 2022 5 commits
  47. 10 Feb, 2022 1 commit
  49. 04 Feb, 2022 1 commit
  51. 03 Feb, 2022 2 commits
  53. 02 Feb, 2022 1 commit
  55. 01 Feb, 2022 2 commits