Hugging Face releases Phonon on-device TTS models

huggingface.co6 min read

Developers have launched Phonon, a small, on-device text-to-speech model available in 65 MB and 75 MB sizes that operates without a network or server. Built on the Pocket TTS lineage using flow matching, Phonon supports five languages and runs locally across phones, laptops, and browsers while topping Hugging Face's Open TTS Leaderboard for CPU streaming speed and intelligibility. Open article →