Name
Abstract | ||
|---|---|---|
Though depression is a common mental health problem with significant impact on human society, it often goes undetected. We explore a diverse set of features based only on spoken audio to understand which features correlate with self-reported depression scores according to the Beck depression rating scale. These features, many of which are novel for this task, include (1) estimated articulatory trajectories during speech production, (2) acoustic characteristics, (3) acoustic-phonetic characteristics and (4) prosodic features. Features are modeled using a variety of approaches, including support vector regression, a Gaussian backend and decision trees. We report results on the AVEC-2014 depression dataset and find that individual systems range from 9.18 to 11.87 in root mean squared error (RMSE), and from 7.68 to 9.99 in mean absolute error (MAE). Initial fusion brings further improvement; fusion and feature selection work is still in progress. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1145/2661806.2661818 | AVEC@MM |
Keywords | Field | DocType |
robust signal analysis,prosody,depression,signal processing,waveform analysis,support vector regression,acoustic features,time series prediction,decision trees,articulatory features | Time series,Prosody,Decision tree,Feature selection,Support vector machine,Psychology,Mean squared error,Rating scale,Speech recognition,Speech production | Conference |
Citations | PageRank | References |
16 | 0.65 | 29 |
Authors | ||
7 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Vikramjit Mitra | 1 | 299 | 24.83 |
| Elizabeth Shriberg | 2 | 3057 | 325.64 |
| Mitchell McLaren | 3 | 454 | 35.97 |
| Andreas Kathol | 4 | 68 | 11.86 |
| Colleen Richey | 5 | 118 | 10.91 |
| Dimitra Vergyri | 6 | 373 | 36.97 |
| Martin Graciarena | 7 | 281 | 24.70 |