pytorch / pytorch/audio

Add pretrained weights from Voxpopuli

Open
#1,920 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
2.9k
Forks
799
Avg merge
58m
Merged PRs (30d)
3

Description

VoxPopuli publishes pre-trained models of many different languages under CC BY-NC 4.0 license. We can add them to torchaudio.

non-fine-tuned weights

https://github.com/facebookresearch/voxpopuli#wav2vec-20

  • es - base
  • es - large
  • fr - base
  • fr - large
  • it - base
  • it - large
  • ni - base
  • ni - large
  • sv - base
  • sv - large
  • 23 langs (10k subset) - base
  • 23 langs (10k subset) - large
  • 23 langs (100k subset) - base
  • 23 langs (100k subset) - large

Fine-tuned ASR

https://github.com/facebookresearch/voxpopuli#asr-and-lm

  • cs
  • de #1953
  • en #1956
  • es #1924
  • et
  • fi
  • fr #1919
  • hr
  • hu
  • it #1954
  • lt
  • ni
  • pl
  • ro
  • sk
  • sl

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the VoxPopuli wav2vec 2.0 and ASR/LM links in the issue, then review how torchaudio currently exposes pretrained models. The checklist is the completion target: add the remaining non-fine-tuned and fine-tuned language weights, while accounting for the already referenced issues #1919, #1924, #1953, and #1954.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.