Feature Augmenting Networks for Improving Depression Severity Estimation From Speech Signals
Open Access
- 30 January 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 24033-24045
Abstract
Depression disorder has become one of the major psychological diseases endangering human health. Researcher in the affective computing community is supporting the development of reliable depression severity estimation system, from multiple modalities (speech, face, text), to assist doctors in their diagnosis. However, the limited amount of annotated data has become the main bottleneck restricting the study on depression screening, especially when deep learning models are used. To alleviate this issue, in this work we propose to use Deep Convolutional Generative Adversarial Network (DCGAN) for features augmentation to improve depression severity estimation from speech. To the best of our knowledge, this approach is the first attempt to apply the Generative Adversarial Network for depression severity estimation from speech. Besides, to measure the quality of the augmented features, we propose three different measurement criteria, characterizing the spatial, frequency and representation learning of the augmented features. Finally, the augmented features are used to train depression estimation models. Experiments are carried out on speech signals from the Audio Visual Emotion Challenge (AVEC2016) depression dataset, and the relationship between the model performance and data size is explored. Our experimental results show that: 1) The combination of the three proposed evaluation criteria can effectively and comprehensively evaluate the quality of the augmented features. 2) When increasing the size of the augmented data, the performance of depression severity estimation gradually improves and the model converges to a certain stable state. 3) The proposed DCGAN based data augmentation approach effectively improves the performance of depression severity estimation, with the root mean square error (RMSE) reduced to 5.520 and mean absolute error (MAE) reduced to 4.634, which is better than most of the state of the art results on AVEC 2016.Keywords
Funding Information
- Shaanxi Provincial International Science and Technology Collaboration Project (2017KW-ZD-14)
- VUB Interdisciplinary Research Program through the EMO-App project
This publication has 40 references indexed in Scilit:
- Multimodal and Multiresolution Depression Detection from Speech and Facial Landmark Features2016
- DepAudioNet2016
- Detecting Depression using Vocal, Facial and Semantic Communication Cues2016
- Depression Assessment by Fusing High and Low Level Features from Audio, Video, and Text2016
- Spectral–Spatial Feature Extraction for Hyperspectral Image Classification: A Dimension Reduction and Deep Learning ApproachIEEE Transactions on Geoscience and Remote Sensing, 2016
- Lung cancer detection using frequency‐domain microwave imagingElectronics Letters, 2015
- Finding the Secret of Image Saliency in the Frequency DomainIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- Multimodal Prediction of Affective Dimensions and Depression in Human-Computer Interactions2014
- AVEC 20142014
- A comparative study of different classifiers for detecting depression from spontaneous speech2013