neuphonic/neutts
★ 6,315 stars
Covered in week 2026-W12 · the 2026-03 report
On-device TTS model by Neuphonic
What neutts is
State-of-the-art Voice AI has been locked behind web APIs for too long. NeuTTS is a collection of open source, on-device, TTS speech language models with instant voice cloning. Built off of LLM backbones, NeuTTS brings natural-sounding speech, real-time performance, built-in security and speaker cloning to your local device - unlocking a new category of embedded voice agents, assistants, toys, and compliance-safe apps.
From the repository README · read it on GitHub →
Transcript
Want natural text to speech running entirely on your phone? New TTS does exactly that. This open-source model generates ultra-realistic voices with instant cloning from just 3 seconds of audio. It runs completely on device on phones, laptops, even Raspberry Pis. No cloud APIs, no latency, full privacy. With 5,000 stars, it's the real deal for embedded voice apps. Star it if you found this helpful. Follow for daily tech finds.
Auto-generated captions — a product name may be transcribed imprecisely.