Model research and engineering lab at @swmansion. We develop, optimize and deploy models for local, edge and hybrid AI, and build agentic systems for business.
We just released Reverb-TTS, a zero-shot, state-of-the-art voice cloning model, made for local inference at @swmansion_ai
Reverb can clone your voice at 17x realtime on Macbook M4 Pro, fully locally in WebGPU via ruNNtime, our own TypeGPU based inference engine π
Obviously, itβs open-source. Go and play with Reverb and show us what you build with it π£οΈ
Demo ππ»
Introducing Reverb-TTS, a state-of-the-art voice-cloning text-to-speech model for local inference. π
Reverb-TTS can clone your voice from one short clip and make it speak Polish, German, French, Spanish, Italian, Portuguese, & English, all on your phone.
It is SotA in both word error rate and speed ποΈ
Itβs the first model from Software Mansion AI Labs, out now in react-native-executorch and ruNNtime.
Run it on your Mac with Nativ, from Neywa Labs @Nativ_AI.
β‘οΈ https://t.co/AwDy3j0671