PREVIOUS
✖
Microsoft’s VASA-1
April 30 , 2024
21 days
93
0
- Microsoft recently introduced a new image-to-video model called VASA-1.
- VASA-1, as an AI model, produces ‘lifelike audio-driven talking faces generated in real time’.
- ‘VAS’ in the name stands for visual affective skill.
- The model is capable of handling types of photos and audio inputs that were not in the training dataset.
Leave a Reply
Your Comment is awaiting moderation.