Automatic Speech Recognition
Laya
Safetensors
speech-to-text
disfluency-removal
punctuation
token-classification
whistle
cactus
mmbert
Instructions to use Lucasgitguy-co/Fleet-Clean-0.36B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Laya
How to use Lucasgitguy-co/Fleet-Clean-0.36B with Laya:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Download tagger/labels.json from Lucasgitguy-co/Fleet-Clean-0.36B: direct link, hf CLI and curl.
- Browser
- Download file 994 Bytes
-
https://huggingface.co/Lucasgitguy-co/Fleet-Clean-0.36B/resolve/main/tagger/labels.json
- Command line
-
hf download hf://Lucasgitguy-co/Fleet-Clean-0.36B/tagger/labels.json
-
curl -L -o labels.json https://huggingface.co/Lucasgitguy-co/Fleet-Clean-0.36B/resolve/main/tagger/labels.json
994 Bytes
| { | |
| "delete": [ | |
| "keep", | |
| "delete" | |
| ], | |
| "punct": [ | |
| "", | |
| ",", | |
| ".", | |
| "?", | |
| "!" | |
| ], | |
| "case": [ | |
| "keep", | |
| "upper_first", | |
| "lower_first" | |
| ], | |
| "layout": "logits[0:2] delete, [2:7] punct, [7:10] case", | |
| "label_token": "first sub-token of each whitespace-separated word", | |
| "step": 528, | |
| "validation": { | |
| "delete_f1": 0.928709055876686, | |
| "delete_precision": 0.9168478260869565, | |
| "delete_recall": 0.9408812046848857, | |
| "false_delete_rate": 0.004597355769230769, | |
| "punct_macro_f1": 0.7164910954761785, | |
| "case_acc": 0.988927960395813, | |
| "score": 0.8780427039162259, | |
| "step": 528 | |
| }, | |
| "base": "convaiinnovations/laya-multilingual encoder", | |
| "delete_threshold": 0.5, | |
| "delete_threshold_rule": "smallest of 0.5..0.95 with <= 0.5% of fluent validation words deleted in every language", | |
| "punct_mode": "boundary", | |
| "punct_mode_rule": "chosen on DisfluencySpeech validation (scripts/punct_modes.py): full re-punctuation scored below Fleet-2's own marks there; 'full' is available as an option" | |
| } |