Fig. 4From: Multi-rate modulation encoding via unsupervised learning for audio event detectionClasswise performance breakup indicated by PSDS scoresBack to article page