Preprint
Aug 2026
AudioSpan: Spanning the Duration and Depth of Audio Comprehension
This work introduces AudioSpan, a benchmark that spans both duration and depth: it pairs audio from 10 minutes to over 2 hours with 3,240 questions across three cognitive levels, namely perception, understanding, and reasoning.
Wen Huang, Yunfei Chu, Meng Gao et al.
· 0 citations