Return
Company Page
Miso Labs
Founders
Miso Labs is building the world's most emotive foundation models for voice, empowering developers to create voice agents that feel genuinely human. Their flagship model, MisoTTS, is an 8-billion-parameter transformer that generates highly expressive speech from both text and audio context, achieving just 110 milliseconds of latency. The company offers real-time latency, one-shot voice cloning, and on-premises deployment, with open-source model weights available on Hugging Face.
Company Founded in 2025
SaaS