Underdog shows off a 32MB speech model in new demo

Despite its small size, Underdog's new model can generate both speech and music.

TReported by TechCrunch. We will add other newspapers as they cover it. November 6, 2026, 13:02 UTC

How the newspaper covers this story

1. TTechCrunchUnited StatesNovember 6, 2026, 13:02 UTC

   Underdog shows off a 32MB speech model in new demo

   Despite its small size, Underdog's new model can generate both speech and music.

   Read at TechCrunchRead translated