126 likesNew · 5d old
Tiny CPU-only TTS model beats big audio giants
What is Underdog-Woof?
Underdog Woof is a 40 MB English text-to-speech model that runs on a laptop CPU using only NumPy. It offers two voices, reads numbers and dates naturally, and is over a thousand times smaller than the 41 GB Fish Audio Drama 3 model.
Who it is for
If you need quick, low-resource speech synthesis for scripts or accessibility, you can run it locally without a GPU or cloud API.
How to try it
Open the demo at https://underdog.ai/woof
The story so far
Early Signal first spotted ConwayResearch/Underdog-Woof on November 9, 2026, with 126 likes. On its busiest day it gained 25 likes. The project is 5 days old. Early Signal files it under Voice, Inference, RAG & Memory.
Questions about Underdog-Woof
- How do I try Underdog-Woof?
- Open the demo at https://underdog.ai/woof
- What do I need to run Underdog-Woof?
- Python with NumPy installed