Design of Chinese Depression Spoken Speech Corpus Based on Psychological Self-related Processing

Abstract

In recent years, speech-based depression detection has gained some initial interest in China, but the lack of publicly available Chinese depression corpus leads to the difficulty of comparing its features and results. According to psychological theory, the pervasive hopelessness and high risk of suicide in depressive patients suggested that the self might be abnormal among them. It highlights many of the advantages of self-related processing can contribute to corpus design. In this paper, a novel design idea of Chinese depression spoken speech corpus based on psychological self-related processing is outlined in this paper. To do so, we therefore introduce some novel methods such as different psychological self and psychological experimental paradigms. The data and collection method have been used in our study.