JUST IN　•　TechCrunch

Underdog shows off a 32MB speech model in new demo

SHARE THIS STORYPLUS UNLOCKS FAVORITES READ LATER AND SOURCE CONTROL

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

READ ORIGINAL REPORT ↗

THIS JUST HAPPENED PACKAGES THE WORLD INTO A FAST LIVE FEED SOURCE REPORTING STAYS ONE CLICK AWAY

RELATED POSTSMORE FASHION

FASHION　•　news – ZDNET

Thinking Machines’ new model takes the lead in western open-weight AI

FASHION　•　GQ

Our 11 Favorite Watches Coming to the Biggest Auctions of the Year

FASHION　•　GQ

21 Coats and Jackets I Think Are Cool Just Because

FASHION　•　Latest Content - Men's Health

The 8 Best Sweater Polos for Men, According to Style Editors

FASHION　•　GQ

With the ‘New’ Yeezy, Ye Says He Wants to ‘Let the Clothes Do the Talking.’ But Who’s Listening?

FASHION　•　GQ

Men Are Outspending Women on Botox. Here’s What They’re Asking For

FASHION　•　Latest Content - Women's Health

9 Best Winter Boots for Women, Tested and Reviewed by Fashion Editors

FASHION　•　GQ

18 Best Raincoats for Men 2026, According to GQ Editors