A pitch-synchronous single frequency filtering spectrogram improves four-class speech emotion recognition on IEMOCAP to 63.95% unweighted accuracy, with happy emotion recall rising from 0% to 22.7%.
Efficient emotion recognition from speech using deep learning on spectrograms
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Pitch-Synchronous Single Frequency Filtering Spectrogram for Speech Emotion Recognition
A pitch-synchronous single frequency filtering spectrogram improves four-class speech emotion recognition on IEMOCAP to 63.95% unweighted accuracy, with happy emotion recall rising from 0% to 22.7%.