Instructions to use KlangAI/pianissimo-sv-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use KlangAI/pianissimo-sv-mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download KlangAI/pianissimo-sv-mlx --local-dir pianissimo-sv-mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Commit History
23 days ago
-
Wrap code examples at 80 characters
Claude-Session: https://claude.ai/code/session_01JeoiSkEdCzw16SqBenVs8we4c0116 -
Add custom vocabulary (phrase boosting)
transcribe(..., boost=[...]) adds phrase boosting: our implementation of NeMo's phrase boosting (phrase_boost.py), using the model's tokenizer. Claude-Session: https://claude.ai/code/session_01JeoiSkEdCzw16SqBenVs8w7c6550e
26 days ago
-
Split long recordings into overlapping chunks
transcribe() now splits audio longer than chunk_duration (default 120 s) into overlapping pieces (15 s overlap) and joins them with parakeet-mlx's merge, so memory stays flat on long recordings. chunk_duration=None keeps one pass. Claude-Session: https://claude.ai/code/session_01JeoiSkEdCzw16SqBenVs8w2bc7408 -
Add MLX version of Klang Pianissimo
Verified signature9de0373 -
initial commit
Verified signature536ca3e