Home/Events/Underdog Launches 32MB Speech Model Woof for Edge Devices

Underdog Launches 32MB Speech Model Woof for Edge Devices

Confirmed
Confidence
90%
Impact: 70%
Updated 11m ago

Consensus Brief

Underdog has introduced a new 32MB speech model named Woof, designed to run on edge devices without requiring an internet connection. The model features a browser-based game that allows users to interact with a typewriter and music tuner, showcasing its text-to-speech capabilities.

Sourced from
Primary: TechCrunch

What Changed Since Last Update

11m ago

New corroborating source added: TechCrunch published an update on Fri, 06 No ("Underdog shows off a 32MB speech model in new demo").

Claim Ledger

4 claims tracked across sources

Confirmed Fact

Woof is a 32MB speech model with 31 million parameters.

Confirmed Fact

The model can run on most devices without any internet connection.

Confirmed Fact

Underdog also launched a 167MB speech-to-text model named Bark.

Confirmed Fact

The models are open for everyone to download.

Role-Based Impact Analysis