F-Score: 0.68Error Rate: 0.389SED-score: 0.638
Papers with Code · CC BY-SA 4.0
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.
<span class="description-source">Source: A report on sound event detection with different binaural features </span>
논문
PSDS1 (-5dB): 0.049PSDS1 (0dB): 0.114PSDS1 (5dB): 0.175PSDS1 (10dB): 0.212
Rank-1 Recognition Rate: 0.975
PSDS1 (-5dB): 0.134PSDS1 (0dB): 0.219PSDS1 (5dB): 0.291PSDS1 (10dB): 0.356
PSDS1: 0.577
event-based F1 score: 55.6PSDS1: 0.474PSDS2: 0.698
event-based F1 score: 63.4PSDS1: 0.583PSDS2: 0.810
event-based F1 score: 54.0PSDS1: 0.452PSDS2: 0.670
event-based F1 score: 49.6PSDS1: 0.4336PSDS2: 0.8161
event-based F1 score: 50.7
event-based F1 score: 40.7
PSDS1: 0.584
PSDS1: 0.587PSDS2: 0.896
event-based F1 score: 49.62PSDS1: 0.4395PSDS2: 0.6711
event-based F1 score: 25.8
event-based F1 score: 39.0