Automatic Speech Recognition
WhisperKit
Core ML
Portuguese
Spanish
English
macos
apple-silicon
swift
live-captions
subtitles
speech-recognition
machine-translation
on-device
local-first
ollama
Instructions to use Veronyka/livecaption with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- WhisperKit
How to use Veronyka/livecaption 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
Download glossary.example.json from Veronyka/livecaption: direct link, hf CLI and curl.
- Browser
- Download file 709 Bytes
-
https://huggingface.co/Veronyka/livecaption/resolve/main/glossary.example.json
- Command line
-
hf download hf://Veronyka/livecaption/glossary.example.json
-
curl -L -o glossary.example.json https://huggingface.co/Veronyka/livecaption/resolve/main/glossary.example.json
709 Bytes
| { | |
| "_comentario": "Copie para <pasta de dados>/glossary.json (padrão: ~/Library/Application Support/LiveCaption). Reinicie o app. Chaves que começam com _ são ignoradas. pre_substitutions: regex aplicada ao texto falado ANTES de traduzir ($1 = grupo capturado). post_corrections: regex aplicada ao que o Whisper ouviu (corrige nomes que ele erra). keep_terms: termos que o tradutor deve manter como estão. whisper_vocab: deixe vazio (medido: deixa o Whisper ~2x mais lento).", | |
| "whisper_vocab": [], | |
| "post_corrections": [], | |
| "keep_terms": [], | |
| "pre_substitutions": [ | |
| { "pattern": "\\b[Aa] gente\\b", "replace": "nós" }, | |
| { "pattern": "\\b[Ll]egenda(s?)\\b", "replace": "subtítulo$1" } | |
| ] | |
| } | |