Instructions to use argmaxinc/ttskit-coreml with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- WhisperKit
How to use argmaxinc/ttskit-coreml with WhisperKit:
# Install CLI with Homebrew on macOS device brew install whisperkit-cli # View all available inference options whisperkit-cli transcribe --help # Download and run inference using whisper base model whisperkit-cli transcribe --audio-path /path/to/audio.mp3 # Or use your preferred model variant whisperkit-cli transcribe --model "large-v3" --model-prefix "distil" --audio-path /path/to/audio.mp3 --verbose
- Notebooks
- Google Colab
- Kaggle
Add W8A16-stream-multifunction SpeechDecoder
#3
by EduardoPacheco - opened
Adds the streaming SpeechDecoder (ring KV cache, sliding-window attention) for 12hz-0.6b-customvoice and 12hz-1.7b-customvoice. It is faster than W8A16-multifunction across devices and runs fully on the ANE on A14/A15. Existing W8A16 and W8A16-multifunction assets are unchanged.