Gradium Open-Sources Phonon, a 31M-Parameter On-Device TTS Model Under Apache 2.0
MarkTechPost Asif Razzaq
Gradium open-sourced Phonon, an on-device text-to-speech model family with English 31M and other variants. The English Phonon 31m is 31M parameters and ships as a 64.5 MB q8 GGUF file under Apache 2.0. Developers can run TTS locally without a cloud call, with no GPU requirement, and with reported CPU first-audio latency of 6.9 ms.
Why it matters
Gradium has open-sourced Phonon, a family of on-device text-to-speech models. The English Phonon 31m ships under Apache 2.0, fits in 64.5 MB, and starts speaking in about 7 ms on a laptop CPU. The post Gradium Open-Sources Phonon, a 31M-Parameter On-Device TTS Model Under Apache 2.0 appeared first on MarkTechPost.
Related stories
Gradium AI Releases New Default TTS Model: 81.0% Hard-Case Pass Rate at 216 ms Time-to-First-Audio
MarkTechPost · 1 month ago ·
46
Gemini 3.1 Flash TTS: the next generation of expressive AI speech
Google DeepMind · 6 months ago ·
10
Fish Audio Releases Drama 3, a 35B Open-Weight MoE Text-to-Speech Model
MarkTechPost · 1 week ago ·
36