| title | Speaker-Aware Temporal Aggregation Strategies on Segment Representations for Depression Detection in Dyadic Interaction: A Benchmark Study |
|---|---|
| authors | Anisha Pattanayak, Huang-Cheng Chou, Shrikanth Narayanan, Sudarsana Reddy Kadiri |
| date | 2026-07-03 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2607.02904 |
| upvotes | 1 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Speech-based depression detection compresses features from short audio segments into one speaker-level decision, a step called temporal aggregation rarely studied on its own. Most benchmarks fix a single self-supervised encoder and a single hand-picked layer, so a reported gain may reflect the pipeline rather than the aggregation method itself. We introduce DEPOOL, a controlled benchmark that compares six aggregation architectures with six frozen speech backbones on an English and a Mandarin depression corpus, where each configuration learns which backbone layers matter rather than fixing one by hand. Across the resulting 72-configuration grid, a third of configurations collapse into predicting a single class for every speaker, a failure tied to the backbone as much as to the method, and the architecture that is most stable in a single-seed run becomes unreliable when training repeats across seeds. Robustness to backbone and seed, rather than average accuracy across a single pipeline, should be a first-class benchmarking criterion for temporal aggregation in clinical speech. |
cite
@misc{pattanayak2026speaker,
title={Speaker-Aware Temporal Aggregation Strategies on Segment Representations for Depression Detection in Dyadic Interaction: A Benchmark Study},
author={Anisha Pattanayak and Huang-Cheng Chou and Shrikanth Narayanan and Sudarsana Reddy Kadiri},
year={2026},
eprint={2607.02904},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2607.02904}
}