Skip to main content
EURASIP Journal on Advances in Signal Processing

Table 4 Performances of single-condition DNN-HMMs on RealData (WER (%))

From: Reverberant speech recognition combining deep neural networks and deep autoencoders augmented with a phone-class feature

 

RealData

 

Room 1

 

Near (= 100 cm)

Far (= 250 cm)

DNN-HMM (mc)

28.59

30.87

DNN-HMM (Large Near (= 50 cm))

37.18

39.13

DNN-HMM (Large Far (= 200 cm))

28.71

27.95

Back to article page