SOTAVerified

Sound Event Detection

Sound Event Detection (SED) is the task of recognizing the sound events and their respective temporal start and end time in a recording. Sound events in real life do not always occur in isolation, but tend to considerably overlap with each other. Recognizing such overlapping sound events is referred as polyphonic SED.

Source: A report on sound event detection with different binaural features

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 194 papers

TitleStatusHype
Frequency Dynamic Convolutions for Sound Event Detection—0
Hybrid Disagreement-Diversity Active Learning for Bioacoustic Sound Event DetectionCode0
Exploring the Potential of SSL Models for Sound Event Detection—0
Temporal Attention Pooling for Frequency Dynamic Convolution in Sound Event DetectionCode0
Formula-Supervised Sound Event Detection: Pre-Training Without Real Data—0
Exploring Performance-Complexity Trade-Offs in Sound Event Detection ModelsCode1
Aggregation Strategies for Efficient Annotation of Bioacoustic Sound Events Using Active Learning—0
Robust detection of overlapping bioacoustic sound events—0
Synthetic data enables context-aware bioacoustic sound event detection—0
JiTTER: Jigsaw Temporal Transformer for Event Reconstruction for Self-Supervised Sound Event DetectionCode0
Show:102550
← PrevPage 1 of 20Next →
All datasetsDESEDL3DAS21WildDESEDMivia Audio EventsMivia Road Events

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1CRNN (with BEATs + Separation)PSDS1 (-5dB)0.13—Unverified
2CRNN (with BEATs)PSDS1 (-5dB)0.07—Unverified
3CRNN (WildDESED + Curriculrm learning)PSDS1 (-5dB)0.05—Unverified
4CRNN (WildDESED)PSDS1 (-5dB)0.05—Unverified
5CRNNPSDS1 (-5dB)0.02—Unverified