319 likesNew · 5d old
Tiny-voca: 4.5 M-parameter English TTS runs on a phone
What is tiny-voca?
tiny-voca is a tiny English text-to-speech model that offers a female and a male voice. At only 4.5 M parameters it can synthesize speech about 70× faster than real time on four CPU threads.
Who it is for
Developers who need lightweight, low-latency speech synthesis for mobile or browser apps should care.
How to try it
Open the demo at https://huggingface.co/spaces/n4ze3m/tiny-voca-studio
The story so far
Early Signal first spotted n4ze3m/tiny-voca on November 19, 2026, with 319 likes. On its busiest day it gained 63 likes. The project is 5 days old. Early Signal files it under Voice, Coding.
Questions about tiny-voca
- How do I install tiny-voca?
- Run: pip install git+https://github.com/n4ze3m/dizione git+https://github.com/n4ze3m/tiny-voca
- What do I need to run tiny-voca?
- Python with pip