Hugging Face launches Phonon TTS, on-device speech model tops benchmarks

Phonon is a 31 million‑parameter text‑to‑speech model that runs entirely on the user’s device, requiring no server or network connection. The model’s weights are open‑source (Apache 2.0 for the 31 M version, CC BY‑NC for others) and it achieved top scores on Hugging Face’s Open TTS Leaderboard after a six‑month development cycle following its April announcement.