Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Audio Question Answering research benchmarks – Page 2 · SOTA2 Research

Audio Question Answering

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

MMAU mini v05.15.25 (test)

76.8Accuracy

8

Jun 9, 2026

speech-chatbot-alpaca-eval (S-Alpaca)

66.7Accuracy

8

Apr 10, 2026

85.6Score

7

Feb 26, 2026

Audio Flamingo 3

73.1Overall Score

6

Jul 1, 2026

ST-AudioQA Type-C 1.0 (test)

PSELDNets-mACCDOA + OLMo2

86.3Temp. Rel.

6

Jun 15, 2026

TELEVAL AQA-en (dev)

57.69TELEVAL Score

6

Feb 26, 2026

TELEVAL AQA-zh (dev)

49.95TELEVAL Score

6

Feb 26, 2026

Industrial-IoT Simple-QA n=800 (test)

90.07Detection Accuracy

4

Feb 26, 2026

Home-IoT Simple-QA n=800 (test)

90.67Detection Accuracy

4

Feb 26, 2026

59.17M.J. Score

3

Feb 26, 2026

46.5Accuracy

2

Feb 26, 2026

ClothoAQA numerical

36.4Accuracy

2

Feb 26, 2026

ClothoAQA non-binary

49.5Accuracy

2

Feb 26, 2026

ClothoAQA unanimous

86.9Accuracy

2

Feb 26, 2026

86.2Accuracy

2

Feb 26, 2026

RACE-Audio (test)

—

—Accuracy (Sub-task 1)

0

Feb 26, 2026

Page 2 of 2

Popular datasets

MMAU AudioCaps ClothoAQA Clotho RACE

Follow for updates

Join SOTA2 Contact