Skip to main content

Table 13 SER on the AlbayzĆ­n 2012 test partition for different systems proposed in the literature compared to our proposed RNN approach

From: Multiclass audio segmentation based on recurrent neural networks for broadcast domain data

System

SER

RNN proposal (pool + mixup)

24.93

DCASE baseline

31.21

GMM + Viterbi decoding [67]

26.34

HMM-GMM [68]

26.53