Saved in:
Bibliographic Details
Main Authors: Harati, Amir, Shriberg, Elizabeth, Rutowski, Tomasz, Chlebek, Piotr, Lu, Yang, Oliveira, Ricardo
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.16900
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917879542185984
author Harati, Amir
Shriberg, Elizabeth
Rutowski, Tomasz
Chlebek, Piotr
Lu, Yang
Oliveira, Ricardo
author_facet Harati, Amir
Shriberg, Elizabeth
Rutowski, Tomasz
Chlebek, Piotr
Lu, Yang
Oliveira, Ricardo
contents Speech-based algorithms have gained interest for the management of behavioral health conditions such as depression. We explore a speech-based transfer learning approach that uses a lightweight encoder and that transfers only the encoder weights, enabling a simplified run-time model. Our study uses a large data set containing roughly two orders of magnitude more speakers and sessions than used in prior work. The large data set enables reliable estimation of improvement from transfer learning. Results for the prediction of PHQ-8 labels show up to 27% relative performance gains for binary classification; these gains are statistically significant with a p-value close to zero. Improvements were also found for regression. Additionally, the gain from transfer learning does not appear to require strong source task performance. Results suggest that this approach is flexible and offers promise for efficient implementation.
format Preprint
id arxiv_https___arxiv_org_abs_2412_16900
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Speech-Based Depression Prediction Using Encoder-Weight-Only Transfer Learning and a Large Corpus
Harati, Amir
Shriberg, Elizabeth
Rutowski, Tomasz
Chlebek, Piotr
Lu, Yang
Oliveira, Ricardo
Audio and Speech Processing
Computation and Language
Speech-based algorithms have gained interest for the management of behavioral health conditions such as depression. We explore a speech-based transfer learning approach that uses a lightweight encoder and that transfers only the encoder weights, enabling a simplified run-time model. Our study uses a large data set containing roughly two orders of magnitude more speakers and sessions than used in prior work. The large data set enables reliable estimation of improvement from transfer learning. Results for the prediction of PHQ-8 labels show up to 27% relative performance gains for binary classification; these gains are statistically significant with a p-value close to zero. Improvements were also found for regression. Additionally, the gain from transfer learning does not appear to require strong source task performance. Results suggest that this approach is flexible and offers promise for efficient implementation.
title Speech-Based Depression Prediction Using Encoder-Weight-Only Transfer Learning and a Large Corpus
topic Audio and Speech Processing
Computation and Language
url https://arxiv.org/abs/2412.16900