Orpheus-TTS
Canopy AI's open-source high-quality TTS on LLaMA architecture — beats commercial models
Canopy AI's open-source high-quality TTS built on LLaMA architecture, claiming to surpass commercial models; strong emotional expressiveness.
Voice AI: lifelike TTS, accurate ASR, and real-time voice agents.
Key features
- Open-source TTS built on the Llama-3b backbone
- Zero-shot voice cloning without prior fine-tuning
- Emotion and intonation control with simple tags
- About 200ms streaming latency for realtime applications
Good for
- Developers needing low-latency streaming TTS
- Apps adding expressive voice with guided emotion tags
Stack & tags
PythonvLLMLlama.cppttsopen-weightszero-shot-cloningstreamingllm-basedlow-latency
Quick start
git clone https://github.com/canopyai/Orpheus-TTS.git
cd Orpheus-TTS && pip install orpheus-speechFrom the project README on GitHub.