TADA — Text-Acoustic Dual Alignment for Speech
🗯
22
Speech generation from text and acoustic reference
Rebuilt AI Research Demos
Research demos rebuilt for clarity, exploration, and better understanding of AI models.
Note This Space rebuilds the original TADA demo with a clearer workflow and more guided UI. It highlights the text–acoustic alignment process, adds multilingual support, and makes the full pipeline easier to explore step by step.
-
NVILA HD Video + AutoGaze
🎯9Long video understanding with smart attention