Table 1.
Recall, precision, and F1-score in the binary case.
Table 2.
Emotions considered in bilingual emotion recognition with a common model set.
Fig 1.
Computation of shifted delta cepstral (SDC) coefficients.
Fig 2.
Architecture of the proposed convolutional neural networks-based classifier.
Table 3.
Spoken language identification rates [%] using English and German emotional speech data.
Table 4.
Recalls for speech emotion recognition using IEMOCAP and DNN.
Table 5.
Recalls for speech emotion recognition using IEMOCAP and CNN.
Table 6.
Precision of speech emotion recognition using IEMOCAP and DNN.
Table 7.
Precision of speech emotion recognition using IEMOCAP and CNN.
Table 8.
F1-scores for speech emotion recognition using IEMOCAP and DNN.
Table 9.
F1-scores for speech emotion recognition using IEMOCAP and CNN.
Table 10.
Confusion matrix [%] using IEMOCAP and DNN with MFCC/SDC features.
Table 11.
Confusion matrix [%] using IEMOCAP and CNN with MFCC/SDC features.
Table 12.
Recalls for speech emotion recognition using FAU Aibo and DNN.
Table 13.
Recalls for speech emotion recognition using FAU Aibo and CNN.
Table 14.
Precision of speech emotion recognition using FAU Aibo and DNN.
Table 15.
Precision of speech emotion recognition using FAU Aibo and CNN.
Table 16.
F1-scores for speech emotion recognition using FAU Aibo and DNN.
Table 17.
F1-scores for speech emotion recognition using FAU Aibo and CNN.
Table 18.
Confusion matrix [%] using FAU Aibo and DNN with MFCC/SDC features.
Table 19.
Confusion matrix [%] using FAU Aibo and CNN with MFCC/SDC features.
Table 20.
Recalls for speech emotion recognition using a common model set and DNN.
Table 21.
Recalls for speech emotion recognition using a common model set and CNN.
Table 22.
Precision of speech emotion recognition using a common model set and DNN.
Table 23.
Precision of speech emotion recognition using a common model set and CNN.
Table 24.
F1-scores for speech emotion recognition using a common model set and DNN.
Table 25.
F1-scores for speech emotion recognition using a common model set and CNN.
Table 26.
Training and test instances for the IEMOCAP corpus.
Table 27.
Confusion matrix [%] of the spoken language identification in the first pass.
Fig 3.
UARs for multilingual and monolingual emotion recognition for three languages.