Automatic Speech Recognition
NeMo
ONNX
Safetensors
GGUF
parakeet_tdt
parakeet
tdt
sherpa-onnx
multilingual
speech-recognition
gabor
fastconformer
Eval Results (legacy)
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
Add Orukeet v0.1.0 inference package and example
Browse files- orukeet-0.1.0-py3-none-any.whl +0 -0
- orukeet-0.1.0.tar.gz +3 -0
- transcribe.py +24 -0
orukeet-0.1.0-py3-none-any.whl
ADDED
|
Binary file (26.9 kB). View file
|
|
|
orukeet-0.1.0.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:65f57199c411bd266dfabf2800957ebf56c65b25de038f6930ce8d89f9c71929
|
| 3 |
+
size 31074
|
transcribe.py
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""Use an installation receipt without downloading anything during transcription."""
|
| 2 |
+
import argparse
|
| 3 |
+
import json
|
| 4 |
+
import sys
|
| 5 |
+
from pathlib import Path
|
| 6 |
+
|
| 7 |
+
from orukeet import Orukeet
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
def main():
|
| 11 |
+
if hasattr(sys.stdout, "reconfigure"):
|
| 12 |
+
sys.stdout.reconfigure(encoding="utf-8")
|
| 13 |
+
parser = argparse.ArgumentParser(description=__doc__)
|
| 14 |
+
parser.add_argument('audio', type=Path, nargs='+')
|
| 15 |
+
parser.add_argument('--installation', type=Path, default=Path('installation.json'))
|
| 16 |
+
args = parser.parse_args()
|
| 17 |
+
config = json.loads(args.installation.read_text(encoding='utf-8-sig'))
|
| 18 |
+
with Orukeet(config['model'], config['runtime'], device=config['device']) as model:
|
| 19 |
+
for path in args.audio:
|
| 20 |
+
print(json.dumps({'file': str(path), **model.transcribe(path)}, ensure_ascii=False))
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
if __name__ == '__main__':
|
| 24 |
+
main()
|