Automatic Speech Recognition
NeMo
ONNX
Safetensors
GGUF
parakeet_tdt
parakeet
tdt
sherpa-onnx
multilingual
speech-recognition
gabor
fastconformer
Instructions to use oruk/orukeet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NeMo
How to use oruk/orukeet with NeMo:
import nemo.collections.asr as nemo_asr asr_model = nemo_asr.models.ASRModel.from_pretrained("oruk/orukeet") transcriptions = asr_model.transcribe(["file.wav"]) - Notebooks
- Google Colab
- Kaggle
Download transformers/qualification.json from oruk/orukeet: direct link, hf CLI and curl.
- Browser
- Download file 2.06 kB
-
https://huggingface.co/oruk/orukeet/resolve/main/transformers/qualification.json
- Command line
-
hf download hf://oruk/orukeet/transformers/qualification.json
-
curl -L -o qualification.json https://huggingface.co/oruk/orukeet/resolve/main/transformers/qualification.json
2.06 kB
| { | |
| "dataset": "Fixed first20 FLEURS validation clips per en/de/es/fr/ru/uk; compatibility sample, not a held-out model benchmark", | |
| "normalization": "NFKC, lowercase, replace non-alphanumeric nonspace with spaces, whitespace tokens", | |
| "device": "Apple M5 Max CPU,8threads,FP32", | |
| "summary": { | |
| "clips": 120, | |
| "words": 2433, | |
| "export_wer": 0.05425400739827373, | |
| "source_nemo_wer": 0.053431976983148374, | |
| "normalized_source_matches": 117, | |
| "audio_s": 1257.82, | |
| "inference_s": 217.64420087897452 | |
| }, | |
| "languages": { | |
| "de_de": { | |
| "clips": 20, | |
| "words": 393, | |
| "export_wer": 0.04071246819338423, | |
| "source_nemo_wer": 0.03816793893129771, | |
| "normalized_source_matches": 18, | |
| "audio_s": 237.72, | |
| "inference_s": 28.99683033389738 | |
| }, | |
| "en_us": { | |
| "clips": 20, | |
| "words": 382, | |
| "export_wer": 0.034031413612565446, | |
| "source_nemo_wer": 0.034031413612565446, | |
| "normalized_source_matches": 20, | |
| "audio_s": 173.38, | |
| "inference_s": 30.397747127106413 | |
| }, | |
| "es_419": { | |
| "clips": 20, | |
| "words": 499, | |
| "export_wer": 0.03006012024048096, | |
| "source_nemo_wer": 0.028056112224448898, | |
| "normalized_source_matches": 19, | |
| "audio_s": 243.96, | |
| "inference_s": 25.191086375154555 | |
| }, | |
| "fr_fr": { | |
| "clips": 20, | |
| "words": 467, | |
| "export_wer": 0.044967880085653104, | |
| "source_nemo_wer": 0.044967880085653104, | |
| "normalized_source_matches": 20, | |
| "audio_s": 188.7, | |
| "inference_s": 35.01004733797163 | |
| }, | |
| "ru_ru": { | |
| "clips": 20, | |
| "words": 373, | |
| "export_wer": 0.0777479892761394, | |
| "source_nemo_wer": 0.0777479892761394, | |
| "normalized_source_matches": 20, | |
| "audio_s": 220.8, | |
| "inference_s": 36.81056591490051 | |
| }, | |
| "uk_ua": { | |
| "clips": 20, | |
| "words": 319, | |
| "export_wer": 0.11912225705329153, | |
| "source_nemo_wer": 0.11912225705329153, | |
| "normalized_source_matches": 20, | |
| "audio_s": 193.26, | |
| "inference_s": 61.237923789944034 | |
| } | |
| } | |
| } | |