Pith. sign in

REVIEW 5 cited by

Sound Event Detection Transformer: An Event-based End-to-End Model for Sound Event Detection

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2110.02011 v3 pith:KLWND6XX submitted 2021-10-05 cs.SD cs.LGeess.AS

Sound Event Detection Transformer: An Event-based End-to-End Model for Sound Event Detection

Zhirong Ye , Xiangdong Wang , Hong Liu , Yueliang Qian , Rui Tao , Long Yan , Kazushige Ouchi This is my paper
classification cs.SD cs.LGeess.AS
keywords detectioneventmodelsoundtransformerend-to-endpredictionsedt
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Sound event detection (SED) has gained increasing attention with its wide application in surveillance, video indexing, etc. Existing models in SED mainly generate frame-level prediction, converting it into a sequence multi-label classification problem. A critical issue with the frame-based model is that it pursues the best frame-level prediction rather than the best event-level prediction. Besides, it needs post-processing and cannot be trained in an end-to-end way. This paper firstly presents the one-dimensional Detection Transformer (1D-DETR), inspired by Detection Transformer for image object detection. Furthermore, given the characteristics of SED, the audio query branch and a one-to-many matching strategy for fine-tuning the model are added to 1D-DETR to form Sound Event Detection Transformer (SEDT). To our knowledge, SEDT is the first event-based and end-to-end SED model. Experiments are conducted on the URBAN-SED dataset and the DCASE2019 Task4 dataset, and both show that SEDT can achieve competitive performance.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Towards Open World Sound Event Detection

    cs.SD 2026-05 unverdicted novelty 7.0

    Introduces OW-SED paradigm and WOOT transformer framework to detect known sounds, identify unseen events, and incrementally learn in open audio environments.

  2. Sound Event Detection with Boundary-Aware Optimization and Inference

    eess.AS 2026-01 conditional novelty 6.0

    Boundary-aware losses plus duration-estimating event proposals raise AudioSet Strong PSDS1 from 46.5 to 49.6 and remove post-processing hyperparameter tuning.

  3. Towards Open World Sound Event Detection

    cs.SD 2026-05 unverdicted novelty 5.0

    Introduces OW-SED paradigm and WOOT framework with deformable attention for detecting known and unseen sound events in open-world settings.

  4. EZhouNet:A framework based on graph neural network and anchor interval for the respiratory sound event detection

    cs.SD 2025-09 reject novelty 5.0

    A GNN plus anchor-interval framework detects abnormal respiratory sound events, reaching F1 of 22.3 percent on SPRSound, with anchor scales tuned to the observed event durations.

  5. A Neuromorphic Trigger for Efficient Audio Event Detection

    cs.SD 2026-06 unverdicted novelty 4.0

    A lightweight fully connected spiking neural network trigger with close-open postprocessing achieves 0.97 F1 on class-agnostic anomalous sound detection and enables 42.6x FLOPs reduction with improved error rate on so...