Instructions to use CHOCHICHO/twirly-ch-gsw-coreml with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- WhisperKit
How to use CHOCHICHO/twirly-ch-gsw-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
Twirly Swiss German (gsw) — CoreML for WhisperKit
CoreML conversion of notebotIE/whisper-large-v2-swiss-german (a Whisper large-v2 fine-tune on the SwissDialDataset_ETH corpus, Apache 2.0) for on-device Swiss German dictation in the Twirly macOS app.
- Transcribes Swiss German dialect audio into written Mundart (SwissDial convention) — not Standard German.
- Converted with whisperkittools
(float16; AudioEncoder + TextDecoder + MelSpectrogram
.mlmodelc). - Runs on Apple Silicon via WhisperKit (~6.7× realtime on M-series, ANE).
- Note: the model can emit
<|zc|>-style dialect marker tokens; consumers should decode withskipSpecialTokens.
Conversion + packaging: CHO CHI CHO GmbH, Zürich. License: Apache 2.0 (same as the base fine-tune). Credit for the fine-tune: notebotIE; base model: OpenAI Whisper large-v2.
- Downloads last month
- 32
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for CHOCHICHO/twirly-ch-gsw-coreml
Base model
openai/whisper-large-v2 Finetuned
notebotIE/whisper-large-v2-swiss-german