Instructions to use RegaLabs/RegaLabs-TTS with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- CosyVoice
How to use RegaLabs/RegaLabs-TTS with CosyVoice:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Commit History
-
Publish Sorani LLM adapter, fix Space default prompt, scrub blocklist terms from docs
- add cosyvoice3_sorani_lora_refined_best.pt (required for Sorani speech) - app.py/infer.py: load the LLM adapter; fix default prompt transcript to match the bundled sample audio - README: updated usage; generic wording for the content filter - censor.py: remove per-term references from docstring/comments5da3b02 -
Add fail-closed Sorani content filter, no-liability license terms, and endofprompt fix
- sorani/censor.py: dictionary-verified sexual-word blocklist (کێر/کیر/قوز/کوز), bleep replacement, SHA-256 wordlist + checkpoint integrity checks - frontend/app/infer: route text through the censoring filter - infer.py: inject <|endofprompt|> (CosyVoice3 requirement) - LICENSE + README: consent requirement, prohibited uses, no-liability disclaimer for misusedee8fe1
-
Update license: Apache 2.0 with RegaLabs credit (commercial) & non-commercial stock voice restriction
56ecd48
-
Upload RegaLabs-TTS model weights, infer.py, and app.py Gradio demo
Verified signaturee352304