Add comprehensive TTS model documentation and research (!1) · Merge requests · engineering / unturf / uncloseai-speech · GitLab

grow food, not lawn.

GitLab now enforces expiry dates on tokens that originally had no set expiration date. Those tokens were given an expiration date of one year later. Please review your personal access tokens, project access tokens, and group access tokens to ensure you are aware of upcoming expirations. Administrators of GitLab can find more information on how to identify and mitigate interruption in our documentation.

Russell Ballestrini requested to merge claude/implement-models-docs-011CUxXuNMytPjEr5vsvcboo into main Nov 09, 2025

Added detailed documentation for 10 TTS models:

Coqui TTS (XTTS-v2): High-quality multilingual with voice cloning
Mozilla TTS: Historical context, superseded by Coqui
Piper TTS: Fast, lightweight, 100+ voices
Chatterbox: Emotion control, 23 languages
Mimic 3: Privacy-focused, offline capable
eSpeak NG: 100+ languages, accessibility-focused
Kokoro TTS: Fast decoder-only architecture
Tortoise TTS: Studio-quality but slow
Step-Audio-EditX: Experimental LLM-based audio editing
Maya1: Indic languages specialist

Created comprehensive research overview:

Complete model comparison matrix
Performance characteristics and feature analysis
License compatibility analysis
Integration roadmap and priorities
Raccoon Mission risk assessment

Updated MODELS.md:

Added documentation index with links to all model docs
Added research overview reference
Added detailed doc references in existing sections
Added "Additional Models Under Research" section

All documentation follows Raccoon Mission theme of rescuing abandoned open-source TTS models for long-term preservation.

All source code donated into the public domain. All trademarks reserved, including: unturf, Remarkbox, Make Post Sell