Who leads the ESC-50 benchmark?

BEATs (iter3+) currently leads ESC-50 with a score of 98.10 on accuracy.

What is the state-of-the-art score on ESC-50?

The state-of-the-art result on ESC-50 is 98.10 (accuracy), achieved by BEATs (iter3+) as of 2022.

How many models are tracked on ESC-50?

Codesota tracks 4 models on ESC-50.

When was the ESC-50 leaderboard last updated?

The ESC-50 leaderboard on Codesota includes results through 2022, with the earliest tracked result from 2021.

Codesota · Audio · Audio Classification · ESC-50Tasks/Audio/Audio Classification

Audio Classification · benchmark dataset · 2015 · N/A

Environmental Sound Classification 50.

Name: Environmental Sound Classification 50 Benchmark Results
Creator: Codesota
Published: 2021-01-01
License: https://creativecommons.org/licenses/by/4.0/

2,000 environmental audio recordings organized into 50 classes (animals, natural soundscapes, etc.).

Paper ↗Submit a result ↵

§ 01 · Leaderboard

Best published scores.

4 results indexed across 1 metric. Shaded row marks current SOTA; ties broken by submission date.

Primary: accuracy · higher is better

accuracy· primary

4 rows

#	Model	Org	Submitted	Paper / code	accuracy
01	BEATs (iter3+)	—	Dec 2022	BEATs: Audio Pre-Training with Acoustic Tokenizers · code	98.10
02	AST-P	—	Apr 2021	AST: Audio Spectrogram Transformer · code	95.60
03	CLAP+K2C Aug.	—	Nov 2022	Large-scale Contrastive Language-Audio Pretraining with … · code	91
04	AST-S	—	Apr 2021	AST: Audio Spectrogram Transformer · code	88.70

Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.

§ 03 · Progress

2 steps
of state of the art.

Each row below marks a model that broke the previous record on accuracy. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Higher scores win. Each subsequent entry improved upon the previous best.

SOTA line · accuracy

Apr 5, 2021AST-P95.60
Dec 18, 2022BEATs (iter3+)98.10

Fig 3 · SOTA-setting models only. 2 entries span Apr 2021 → Dec 2022.

§ 04 · Literature

3 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

BEATs: Audio Pre-Training with Acoustic Tokenizers
Dec 2022·BEATs (iter3+)
arXiv ↗Code
Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation
Nov 2022·CLAP+K2C Aug.
arXiv ↗Code
AST: Audio Spectrogram Transformer
Apr 2021·AST-P, AST-S
arXiv ↗Code

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide

What a submission needs

01A public checkpoint or API endpoint
02A reproduction script with frozen commit + seed
03Declared evaluation environment (Python, deps)
04One row per metric declared by this dataset
05A contact so we can follow up on discrepancies

Environmental Sound Classification 50.

Best published scores.

2 stepsof state of the art.

3 paperstied to this benchmark.

Neighbouring benchmarks.

Have a score that beatsthis table?

2 steps
of state of the art.

3 papers
tied to this benchmark.

Have a score that beats
this table?