Show HN: Pocket TTS running in (mobile) Safari

Category: ai-ml

Tags: text-to-speech, cpu-inference, voice-cloning

Score: 7.3/10 (Innovation: 7, Technical: 7, Documentation: 8, Utility: 7)

Pocket TTS is a lightweight text-to-speech model that runs efficiently on CPUs, supporting multiple languages and voice cloning without requiring GPU hardware. Its innovative combination of small model size (100M parameters), low latency (~200ms), and browser-compatible deployment makes it highly practical for edge and client-side TTS applications.

Target audience: backend devs, data engineers

Repository: https://ldenoue.github.io/xn-ptts/ · Python · MIT · 4465 stars

View on Hacker News