SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

[Wav2Vec2 Speech Event] Add speech event v2 (#15083)

Browse Source 
* up

* up

* up

* up

* up

* up

* improve

* up

* up

* Update src/transformers/trainer.py

* up

* up

* up
This commit is contained in:
Patrick von Platen
2022-01-10 10:46:21 +01:00
committed by GitHub
parent 768e6c1449
commit d72343d2b8
5 changed files with 409 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/research_projects/wav2vec2/README.md
View File

@@ -1,12 +1,12 @@
**NOTE**: This example is outdated and is not longer actively maintained. Please
follow the new instructions of fine-tuning Wav2Vec2 [here](https://github.com/huggingface/transformers/blob/master/examples/pytorch/speech-recognition/README.md)
## Fine-tuning Wav2Vec2
The `run_asr.py` script allows one to fine-tune pretrained Wav2Vec2 models that can be found [here](https://huggingface.co/models?search=facebook/wav2vec2).
This finetuning script can also be run as a google colab [TODO: here]( ).
The script is actively maintained by [Patrick von Platen](https://github.com/patrickvonplaten).
Feel free to ask a question on the [Forum](https://discuss.huggingface.co/) or post an issue on [GitHub](https://github.com/huggingface/transformers/issues/new/choose) and adding `@patrickvonplaten` as a tag.
### Fine-Tuning with TIMIT
Let's take a look at the [script](./finetune_base_timit_asr.sh) used to fine-tune [wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base)
with the [TIMIT dataset](https://huggingface.co/datasets/timit_asr):