# MMAU-Pro: Voice Benchmark Profile

> Tests harder audio reasoning across speech, sound, music, spatial audio, multiple clips, voice chat, and instruction following.

- Scope: 5,305 human-curated questions; 49 skills; audio up to 10 minutes
- Measurement lanes: Spoken reasoning, Conversation dynamics
- Primary metric: Average accuracy plus task-family accuracy
- Available evidence: Machine-readable benchmark-owner leaderboard
- Owner: MMAU-Pro authors
- Source snapshots refreshed: 2026-10-28

## Interpretation limit

The overall average spans heterogeneous audio tasks. Inspect speech, voice, spatial, multi-audio, and instruction-following columns before using it to choose a system.

## Primary sources

- [Paper](https://arxiv.org/abs/2508.13992)
- [Code](https://github.com/sonalkum/MMAUPro)
- [Dataset](https://huggingface.co/datasets/gamma-lab-umd/MMAU-Pro)
- [Owner page](https://sonalkum.github.io/mmau-pro/#results)

Canonical page: https://benchlm.ai/voice-benchmarks/mmau-pro