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 Classification research benchmarks – Page 3 · SOTA2 Research

Audio Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

AudioMarathon 1.0 (test)

Qwen2.5-Omni-7B

78.4SED Score

15

Feb 26, 2026

Ensemble (Submission)

46.3Score

14

Jun 26, 2026

Speech Commands

97.8Accuracy

14

Jun 12, 2026

92.63Accuracy (Session 0)

13

Jul 8, 2026

96.6Anth F1-score

13

May 21, 2026

95Accuracy

13

Mar 17, 2026

Speech Commands (SC) Unprocessed signals (RAW)

98.32Accuracy

13

Feb 26, 2026

AudioSet-20K (test)

37.4mAP

13

Feb 26, 2026

AudioSet (balanced)

37.8mAP

13

Feb 26, 2026

99.87Accuracy (Session 1)

12

Jun 9, 2026

XARES-LLM Track A

Multi-expert Encoder (Supervised)

80.6Track A Score

12

Jun 5, 2026

80.1Top-1 Accuracy

12

May 19, 2026

UrbanSound8K In-Domain

40.85AI

12

Feb 26, 2026

ESC50 (In-Domain)

43.35AI

12

Feb 26, 2026

AudioSet 20K v1

41.9mAP

11

Feb 26, 2026

41.1mAP

11

Feb 26, 2026

50.2mAP

10

Feb 26, 2026

Dataset β with pruning

48.9Accuracy

10

Feb 26, 2026

Dataset β with pruning

91.5Accuracy

10

Feb 26, 2026

70.5Exact-match Accuracy

10

Feb 26, 2026

TUT Urban Acoustic Scenes Mobile (Random pair) 2018 (target)

62.7Average Accuracy

10

Feb 26, 2026

TUT Urban Acoustic Scenes Mobile (1:1 pair) target 2018

85.7Average Accuracy

10

Feb 26, 2026

99Top-1 Acc

10

Feb 26, 2026

FSD50K (train/test)

58.27mAP

9

Jun 1, 2026

78.1Accuracy

9

Apr 30, 2026

Page 3 of 7

Popular datasets

AudioSet Speech Commands ESC UrbanSound TUT

Follow for updates

Join SOTA2 Contact