Instructions to use NeoRoth/oriloq-asr-fr-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use NeoRoth/oriloq-asr-fr-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir oriloq-asr-fr-mlx NeoRoth/oriloq-asr-fr-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
oriloq-asr-fr-mlx
Internal ASR model used by the Oriloq macOS app.
The repository content is updated in place when a new revision is released. Clients should track the repo HEAD commit SHA to detect updates.
License attribution
Fine-tuned LoRA variant of nvidia/parakeet-tdt-0.6b-v3, distributed under CC-BY-4.0.
- Original work: NVIDIA NeMo team.
- Modifications: LoRA fine-tuning by Oriloq for French speech recognition.
When redistributing or presenting this model in a downstream application, retain attribution to NVIDIA, indicate the modification, and reference the CC-BY-4.0 license.
- Downloads last month
- 21
Model size
0.6B params
Tensor type
F32
·
Hardware compatibility
Log In to add your hardware
Quantized
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for NeoRoth/oriloq-asr-fr-mlx
Base model
nvidia/parakeet-tdt-0.6b-v3