Hugging Face has released Phonon, a small, on-device text-to-speech model designed for privacy and offline use. The model, available in two sizes, supports multiple languages and handles complex inputs like phone numbers and dates. Phonon has seen significant adoption, with over 200,000 downloads in its first two weeks, and has also performed well on Hugging Face's Open TTS Leaderboard. AI
IMPACT Enables developers to integrate high-quality, private, offline speech synthesis into applications across various devices.
RANK_REASON This is a product release from a well-known AI platform (Hugging Face) for a specific tool (TTS), but not a frontier model release from a core AI lab.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
