Underdog Demos 32MB Woof Model in a Browser Speech Game
Underdog says the 31-million-parameter model can run on most devices without internet access; the game also features a tuner that generates different kinds of music.

Underdog showcased its 31.7-megabyte text-to-speech model Woof in a browser game posted Nov. 5, using the demo to present speech playback and an interactive music tuner.
Users type on an on-screen typewriter, and Woof reads the entered text aloud. The game’s tuner can be turned to generate different kinds of music. The model has 31 million parameters and is available to download. Underdog says it can run on most devices without an internet connection.
The demo follows Bark, Underdog’s 167-megabyte speech-to-text model for transcription and dictation on edge devices.
TokenPost previously covered a 4-billion-parameter Underdog model named Woof. The Nov. 5 demo describes Woof as a 31-million-parameter text-to-speech model.