TOWARDS LEARNING EMOTION INFORMATION FROM SHORT SEGMENTS OF SPEECH

Conventionally, speech emotion recognition has been approached by utterance or turn-level modelling of input signals, either through extracting hand-crafted low-level descriptors, bag-of-audio-words features, or by feeding long-duration signals directly to deep neural networks (DNNs). While this approach has been successful, there is a growing interest in modeling speech emotion information at short segment level, at around 250ms-500ms (e.g. the 2021-22 MuSe Challenges). This paper investigates both hand-crafted feature-based and end-to-end raw waveform DNN approaches for modeling speech emotion information in such short segments. Through experimental studies on IEMOCAP corpus, we demonstrate that the end-to-end raw waveform modeling approach is more effective than using hand-crafted features for short segment level modelling. Furthermore, through relevance signal-based analysis of the trained neural networks, we observe that the top-performing end-to-end approach tends to emphasize cepstral information instead of spectral information (such as flux and harmonicity).