|
--- |
|
license: apache-2.0 |
|
language: |
|
- ru |
|
library_name: transformers |
|
pipeline_tag: automatic-speech-recognition |
|
tags: |
|
- asr |
|
- Pytorch |
|
- pruned |
|
- audio |
|
- automatic-speech-recognition |
|
metrics: |
|
- cer |
|
- wer |
|
--- |
|
|
|
# Whisper-small-ru-pruned |
|
|
|
## Model info |
|
This is a pruned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) model with only russian tokens left. |
|
Pruning was made without any fine-tuning. Method from [this post](https://medium.com/m/global-identity-2?redirectUrl=https%3A%2F%2Ftowardsdatascience.com%2Fhow-to-adapt-a-multilingual-t5-model-for-a-single-language-b9f94f3d9c90) was used. |
|
|
|
## Size |
|
Only 10% tokens was left including special whisper tokens, added whisper tokens, 100 most popular tokens from tokenizer and 3000 most popular Russian tokens computed by tokenization of russian text corpus. |
|
|
|
Model size is 15% less then original whisper-small: |
|
| | openai/whisper-small | waveletdeboshir/whisper-small-ru-pruned | |
|
| :------ | :------ | :------ | |
|
| n of parameters | 242 M | 205 M | |
|
| n of parameters (with proj_out layer) | 281 M | 209 M | |
|
| model file size | 967 Mb | 837 Mb | |
|
| vocab_size | 51865 | 4705 | |
|
|
|
## Other pruned whisper models |
|
* [waveletdeboshir/whisper-tiny-ru-pruned](https://huggingface.co/waveletdeboshir/whisper-tiny-ru-pruned) |
|
* [waveletdeboshir/whisper-base-ru-pruned](https://huggingface.co/waveletdeboshir/whisper-base-ru-pruned) |
|
|
|
## Metrics |
|
| | openai/whisper-small | waveletdeboshir/whisper-small-ru-pruned | |
|
| :------ | :------ | :------ | |
|
| WER* golos-test-crowd | 0.3358 | 0.3471 | |
|
| CER* golos-test-crowd | 0.1561 | 0.1444 | |
|
*Metrics were measured after text normalization |
|
|
|
You can fine-tune this model on your data to achive better performance. |
|
|
|
## Colab for pruning |
|
TODO |
