Documents
Poster
Multimodal Depression Classification Using Articulatory Coordination Features and Hierarchical Attention Based Text Embeddings
![](https://sigport.org/sites/default/files/styles/home/public/ICASSP_2022_0.jpg?itok=_WnBJxvM)
- Citation Author(s):
- Submitted by:
- Nadee Seneviratne
- Last updated:
- 6 May 2022 - 2:57pm
- Document Type:
- Poster
- Document Year:
- 2022
- Event:
- Presenters:
- Nadee Seneviratne
- Paper Code:
- SPE-9.2
- Categories:
- Keywords:
- Log in to post comments
Multimodal depression classification has gained immense popularity over the recent years. We develop a multimodal depression classification system using articulatory coordination features extracted from vocal tract variables and text transcriptions obtained from an automatic speech recognition tool that yields improvements of area under the receiver operating characteristics curve compared to unimodal classifiers (7.5% and 13.7% for audio and text respectively). We show that in the case of limited training data, a segment-level classifier can first be trained to then obtain a session-wise prediction without hindering the performance, using a multi-stage convolutional recurrent neural network. A text model is trained using a Hierarchical Attention Network (HAN). The multimodal system is developed by combining embeddings from the session-level audio model and the HAN text model.
3649_poster.pdf
![application/pdf PDF icon](/modules/file/icons/application-pdf.png)