ai
Techcrunch Ai
2026-11-06
Underdog demo showcases compact 32MB speech and music generation model
A new AI model from Underdog demonstrates the ability to generate both audio and speech despite an exceptionally small file size.