Worth knowing
Underdog shows off a 32MB speech model in new demo
Ultra-compact, open-weight speech models enable private, offline voice interfaces directly on consumer devices without relying on cloud infrastructure.
Reported by TechCrunch ·
What happened
AI startup Underdog has released Woof, an open 31-million-parameter, 32MB text-to-speech model capable of running locally in a web browser without an internet connection. The release follows the company's 167MB Bark speech-to-text model and a 27-billion-parameter model designed for laptops.
Who it’s about
- Underdog
- Woof
- Bark
- Sigil Wen
Source
More on models
Get a brief picked for you
Sign in to follow topics and companies, save stories and search the archive.
Sign in