Publication

Assessing Bipolar Episodes using Speech Cues derived from Phone Calls

Amir Muaremi, Franz Gravenhorst, Agnes Grünerbl, Bert Arnrich, Gerhard Tröster

In: International Symposium on Pervasive Computing Paradigms for Mental Health (MindCare). International Symposium on Pervasive Computing Paradigms for Mental Health May 8-9 Tokyo Japan Springer Link 5/2014.

Abstract

In this work we show how phone call conversations can be used to objectively predict manic and depressive episodes of bipolar disordered people. In particular, we use phone call statistics, parameters derived from dyadic phone conversations and emotional acoustic features to build and test user-specific classification models. Using random forest, we were able to detect the bipolar states with an average F1 score of 83 %, and we identified the speaking length and phone call length, the HNR value, the number of short turns and the variance of pitch F0 to be the most important variables for prediction

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz