The Effect of Silence Feature in Dimensional Speech Emotion Recognition
This work addresses an incremental improvement in speech emotion recognition for applications like human-computer interaction by exploring the utility of silence features.
The paper investigated whether silence features improve dimensional speech emotion recognition, finding that they affect arousal more than other dimensions and that proper threshold selection boosts performance, while improper selection decreases it, as measured by concordance correlation coefficient.
Silence is a part of human-to-human communication, which can be a clue for human emotion perception. For automatic emotion recognition by a computer, it is not clear whether silence is useful to determine human emotion within a speech. This paper presents an investigation of the effect of using silence feature in dimensional emotion recognition. Since the silence feature is extracted per utterance, we grouped the silence feature with high statistical functions from a set of acoustic features. The result reveals that the silence features affect the arousal dimension more than other emotion dimensions. The proper choice of a threshold factor in the calculation of silence feature improved the performance of dimensional speech emotion recognition performance, in terms of a concordance correlation coefficient. On the other side, improper choice of that factor leads to a decrease in performance by using the same architecture.