← all datasets

10-second V-2A benchmark

Emerging
1papers using it
2024first seen

The '10-second V2A benchmark' is a dataset used to evaluate the performance of video-to-audio generation models specifically for generating audio from video segments that are 10 seconds long.

🔎 Find this dataset

Papers using 10-second V-2A benchmark (1)

10-second V-2A benchmark — datasets — generative-models