Automatic Speech Recognition
Transformers
MLX
English
multilingual
whisper
speech-to-text
quantized
q8
apple-silicon
Instructions to use LibraxisAI/whisper-small-mlx-q8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LibraxisAI/whisper-small-mlx-q8 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="LibraxisAI/whisper-small-mlx-q8")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("LibraxisAI/whisper-small-mlx-q8", dtype="auto") - MLX
How to use LibraxisAI/whisper-small-mlx-q8 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir whisper-small-mlx-q8 LibraxisAI/whisper-small-mlx-q8
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
docs: fix Limitations — audio duration instead of prompt length
Browse files
README.md
CHANGED
|
@@ -101,7 +101,7 @@ No public sample output is currently declared for this checkpoint.
|
|
| 101 |
- No public benchmarks for this checkpoint are declared in the model metadata.
|
| 102 |
- No public benchmark claims are made by this card unless listed in the frontmatter.
|
| 103 |
- Validate outputs on your own domain data before relying on this checkpoint.
|
| 104 |
-
- Memory use and speed depend heavily on
|
| 105 |
|
| 106 |
## License
|
| 107 |
|
|
|
|
| 101 |
- No public benchmarks for this checkpoint are declared in the model metadata.
|
| 102 |
- No public benchmark claims are made by this card unless listed in the frontmatter.
|
| 103 |
- Validate outputs on your own domain data before relying on this checkpoint.
|
| 104 |
+
- Memory use and speed depend heavily on Apple Silicon generation, unified-memory size, audio duration, and language complexity.
|
| 105 |
|
| 106 |
## License
|
| 107 |
|