Automatic Speech Recognition
Transformers
Safetensors
PyTorch
arkasr
text-generation
speech
audio
multilingual
hotword
audio8
custom_code
Eval Results
Instructions to use Edge0/Audio8-ASR-0.1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Edge0/Audio8-ASR-0.1B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Edge0/Audio8-ASR-0.1B", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("Edge0/Audio8-ASR-0.1B", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Fix demo video links
Verified signature8487da6 -
Embed demo video in model card
Verified signatureaf49e11 -
Fix license badge to CC BY-NC 4.0
Verified signaturea74e5f1 -
Update README.md
Verified signature673ef31
-
Report official Open ASR H200 results
Verified signatureUpdate the seven public split WER values and add composite H200 RTFx from the reproducible HF Jobs evaluation.41cda58
-
docs: add German and Korean support
Verified signature9c5becb -
docs: add ASR evaluation results
Verified signature5b6689e -
fix: restore audio attention output layout
Verified signatureRestore the [batch, time, heads, head_dim] layout before flattening multi-head attention output. Local standalone eval10: 11/11 stop, 0/11 max-token hits, 4.73% rough CER. Also remove the temporary README warning.b812eff -
docs: add temporary Transformers release warning
Verified signature7f5c183 -
Public release
25ae981