Tortoise TTS
by James Betker (neonbjb) · open-source · Last verified 2026-03-17
Tortoise TTS is a highly expressive, open-source multi-voice text-to-speech system created by James Betker that achieves exceptional naturalness and zero-shot voice cloning quality through an autoregressive + diffusion pipeline, at the cost of significantly higher inference time than real-time TTS systems. Despite its slow generation speed, Tortoise remains a gold standard for open-source TTS quality and is widely used for offline audiobook and creative narration tasks.
https://github.com/neonbjb/tortoise-tts ↗C+
C+—Average
Adoption: C+Quality: AFreshness: BCitations: B+Engagement: F
Specifications
- License
- Apache 2.0
- Pricing
- open-source
- Capabilities
- text-to-speech, zero-shot-voice-cloning, high-expressiveness, multi-voice, long-form-narration
- Integrations
- huggingface, local-inference
- Use Cases
- audiobook-narration, voice-cloning-research, creative-projects, offline-tts, character-voices
- API Available
- Yes
- Parameters
- ~200M
- Context Window
- N/A
- Modalities
- text, audio
- Training Cutoff
- 2023
- Tags
- text-to-speech, voice-cloning, open-source, high-quality, slow-inference
- Added
- 2026-03-17
- Completeness
- 100%
Index Score
57.6Adoption
55
Quality
88
Freshness
60
Citations
72
Engagement
0