902 likes
Open-weight TTS model streams voice in under 40 ms
What is Breeze-TTS-2?
Breeze-TTS-2 is a free, research-only text-to-speech model that can clone a voice from a short reference, design a new voice from a description, or steer tone and emotion on the fly. It runs in real-time with less than 40 ms time-to-first-audio and a 0.32 real-time factor on an NVIDIA H100.
Who it is for
Developers building chatbots, games, or accessibility tools can generate expressive speech instantly without paying for proprietary APIs.