Sound Event Detection On Wilddesed
평가 지표
PSDS1 (-5dB)
PSDS1 (0dB)
PSDS1 (10dB)
PSDS1 (5dB)
PSDS1 (Clean)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | PSDS1 (-5dB) | PSDS1 (0dB) | PSDS1 (10dB) | PSDS1 (5dB) | PSDS1 (Clean) | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| CRNN (with BEATs + Separation) | 0.134 | 0.219 | 0.356 | 0.291 | 0.440 | Leveraging LLM and Text-Queried Separation for Noise-Robust Sound Event Detection | |
| CRNN (WildDESED) | 0.048 | 0.087 | 0.175 | 0.135 | 0.200 | WildDESED: An LLM-Powered Dataset for Wild Domestic Environment Sound Event Detection System | |
| CRNN (with BEATs) | 0.065 | 0.138 | 0.329 | 0.236 | 0.500 | Leveraging LLM and Text-Queried Separation for Noise-Robust Sound Event Detection | |
| CRNN | 0.017 | 0.064 | 0.222 | 0.148 | 0.348 | WildDESED: An LLM-Powered Dataset for Wild Domestic Environment Sound Event Detection System | |
| CRNN (WildDESED + Curriculrm learning) | 0.049 | 0.114 | 0.212 | 0.175 | 0.265 | WildDESED: An LLM-Powered Dataset for Wild Domestic Environment Sound Event Detection System |
0 of 5 row(s) selected.