Who leads the AudioSet benchmark?

BEATs currently leads AudioSet with a score of 0.51 on map.

What is the state-of-the-art score on AudioSet?

The state-of-the-art result on AudioSet is 0.51 (map), achieved by BEATs as of 2026.

How many models are tracked on AudioSet?

Codesota tracks 4 models on AudioSet.

When was the AudioSet leaderboard last updated?

The AudioSet leaderboard on Codesota includes results through 2026.

Codesota · Benchmark · AudioSetHome/Leaderboards/Audio & Speech/Audio Classification/AudioSet

Unknown

AudioSet.

Name: AudioSet Benchmark Results
Creator: Unknown
Published: 2026-01-01
License: https://creativecommons.org/licenses/by/4.0/

2M+ human-labeled 10-second YouTube video clips covering 632 audio event classes.

Paper ↗Leaderboard ↓Lineage

§ 01 · Leaderboard

Results by metric.

Only 4 models on this benchmark

Help build the community leaderboard — submit your model results.

Found a wrong score or missing run?

Use row edits to send a sourced correction into moderation.

Add / edit result ↗Report issue ↗

map

Map is the reported evaluation metric for AudioSet. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for mapverifiedpapervendorcommunityunverified

Muted rows were not state of the art when published — an earlier or same-year result already scored better.

Rank	Model	Trust	Score	Year	Links	Fix
01	BEATs Fetched from CodeSOTA API on 2026-04-20	verified	0.51	2026	Source ↗	Looks wrong?
02	AST Fetched from CodeSOTA API on 2026-04-20	verified	0.48	2026	Source ↗	Looks wrong?
03	HTS-AT Fetched from CodeSOTA API on 2026-04-20	verified	0.47	2026	Source ↗	Looks wrong?
04	CLAP Fetched from CodeSOTA API on 2026-04-20	verified	0.43	2026	Source ↗	Looks wrong?

Lineage

AudioSet in context.

See full audio understanding benchmarks lineage →

Predecessors (1)

saturated2015-01

ESC-50

AudioSet replaced ESC-50 as the primary audio classification benchmark — 527 classes vs 50, 2M clips vs 2K, hierarchical ontology. Scale and coverage made it the ImageNet analogue for audio. ESC-50 became a probe task for pretrained representations.

This benchmark (1)

saturating2017-03

AudioSet

Successors (1)

active2020-01

Clotho

Clotho shifted the evaluation task from classification (what sounds are here?) to captioning (describe these sounds in a sentence). A scope shift enabled by the growing capability of audio encoders trained on AudioSet.

§ 04 · Submit a result

Add to the leaderboard.

Submit a Result

← Back to Audio Classification