Boson AI targets voice market with Higgs RealTime model supporting 100+ languages

smolix · x · 2026-08-23

Boson AI, founded by ML researcher Alex Smola, is preparing to release its first speech-to-speech model, Higgs RealTime, aiming to revolutionize human-machine interaction. The company's Higgs TTS 3 supports expressive conversational speech in over 100 languages, featuring zero-shot voice cloning and inline emotion control. They also launched the Higgs Avatar API for generating real-time talking-head videos. Previous models were open-sourced with over 10 million hours of training data.

Original post →

More from Models

Models channel →