Papers › Fine-tune the pretrained ATST model for sound event detection

Fine-tune the pretrained ATST model for sound event detection

15 Sep 2023arXiv:2309.08153archive 2025-07-28

Nian Shao, Xian Li, Xiaofei Li

Sound event detection (SED) often suffers from the data deficiency problem. The recent baseline system in the DCASE2023 challenge task 4 leverages the large pretrained self-supervised learning (SelfSL) models to mitigate such restriction, where the pretrained models help to produce more discriminative features for SED. However, the pretrained models are regarded as a frozen feature extractor in the challenge baseline system and most of the challenge submissions, and fine-tuning of the pretrained models has been rarely studied. In this work, we study the fine-tuning method of the pretrained models for SED. We first introduce ATST-Frame, our newly proposed SelfSL model, to the SED system. ATST-Frame was especially designed for learning frame-level representations of audio signals and obtained state-of-the-art (SOTA) performances on a series of downstream tasks. We then propose a fine-tuning method for ATST-Frame using both (in-domain) unlabelled and labelled SED data. Our experiments show that, the proposed method overcomes the overfitting problem when fine-tuning the large pretrained network, and our SED system obtains new SOTA results of 0.587/0.812 PSDS1/PSDS2 scores on the DCASE challenge task 4 dataset.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2309.08153")

Code

Syntology Ran 10 of 10 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 1 ran · fixture could not drive it; 9 ran with no contract checked.

By repository: official repository: 10 samples from 1 repository, 10 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Audio-WestlakeU/ATST-SED officialmentioned in papermentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested; 10 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · fixture could not drive it
9ran

Licence: 0 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from Audio-WestlakeU/ATST-SED. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

drop_path Audio-WestlakeU/ATST-SED/desed_task/nnet/atst/transformer.py official repository ran · fixture could not drive it fingerprinted MIT (permissive) · 55120f2026b56aa2 · report
get_attention_mask Audio-WestlakeU/ATST-SED/desed_task/nnet/atst/transformer.py official repository ran MIT (permissive) · 2e55511eb5179c63 · report
get_cls_avg Audio-WestlakeU/ATST-SED/desed_task/nnet/atst/audio_transformer.py official repository ran MIT (permissive) · 38d5542f89a80703 · report
get_num_patches Audio-WestlakeU/ATST-SED/desed_task/nnet/atst/audio_transformer.py official repository ran fingerprinted MIT (permissive) · 4256112ba8f8c7ab · report
mixup Audio-WestlakeU/ATST-SED/desed_task/data_augm.py official repository ran MIT (permissive) · 4450267615e72051 · report
mixup_w_pretrained Audio-WestlakeU/ATST-SED/desed_task/data_augm.py official repository ran MIT (permissive) · 4b00ec1c23e7e9e8 · report
pad_audio Audio-WestlakeU/ATST-SED/desed_task/dataio/datasets_atst_sed.py official repository ran MIT (permissive) · 6d5765c59e00e650 · report
process_labels Audio-WestlakeU/ATST-SED/desed_task/dataio/datasets_atst_sed.py official repository ran MIT (permissive) · 0cc07361c4eb65c4 · report
to_mono Audio-WestlakeU/ATST-SED/desed_task/dataio/datasets_atst_sed.py official repository ran MIT (permissive) · 01e1fd4bd6d27829 · report
trunc_normal_ Audio-WestlakeU/ATST-SED/desed_task/nnet/atst/audio_transformer.py official repository ran MIT (permissive) · 7810fed39bf5c241 · report

Tasks

Event DetectionSelf-Supervised LearningSound Event Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Sound Event Detection DESED ATST-SED PSDS1 0.583 #1 of 13 Archive leaderboard report
Sound Event Detection DESED ATST-SED PSDS2 0.810 #1 of 13 Archive leaderboard report
Sound Event Detection DESED ATST-SED event-based F1 score 63.4 #1 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections