Instructions to use Tohirju/sl-kestrel with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NeMo
How to use Tohirju/sl-kestrel with NeMo:
# tag did not correspond to a valid NeMo domain.
- Notebooks
- Google Colab
- Kaggle
You need to agree to share your contact information to access this model
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
Access requires approval by the repository owner.
Log in or Sign Up to review the conditions and access this model content.
sl-kestrel
NeMo RNNT/TDT ASR checkpoints.
| file | FLEURS n=600 WER / CER (greedy) |
|---|---|
best.nemo |
11.66 / 4.35 (beam5: 11.45 / 4.31) |
alt_checkpoints/ep5.75.nemo |
11.81 / 4.45 |
alt_checkpoints/ep5.00.nemo |
11.90 / 4.46 |
import nemo.collections.asr as nemo_asr
m = nemo_asr.models.ASRModel.restore_from("best.nemo")
print(m.transcribe(["audio.wav"])[0].text)
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support