The Relationship Between Speech Features Changes When You Get Depressed: Feature Correlations for Improving Speed and Performance of Depression Detection

TL;DR

This paper demonstrates that depression alters speech feature correlations and that leveraging these correlations enhances the training speed and accuracy of depression detection models, with significant error reduction.

Contribution

It introduces the use of feature correlation matrices for depression detection, improving model training speed and accuracy over traditional feature vectors.

Findings

Error rate reduced by up to 26.6% using correlation matrices.

Feature correlations are more variable in depressed speakers, serving as potential depression markers.

Models trained on correlation matrices outperform those trained on feature vectors.

Abstract

This work shows that depression changes the correlation between features extracted from speech. Furthermore, it shows that using such an insight can improve the training speed and performance of depression detectors based on SVMs and LSTMs. The experiments were performed over the Androids Corpus, a publicly available dataset involving 112 speakers, including 58 people diagnosed with depression by professional psychiatrists. The results show that the models used in the experiments improve in terms of training speed and performance when fed with feature correlation matrices rather than with feature vectors. The relative reduction of the error rate ranges between 23.1% and 26.6% depending on the model. The probable explanation is that feature correlation matrices appear to be more variable in the case of depressed speakers. Correspondingly, such a phenomenon can be thought of as a depression marker.