Promoting the Responsible Development of Speech Datasets for Mental Health and Neurological Disorders Research

Download

• Paper
• Code and data

Abstract

Current research in machine learning and artificial intelligence is largely centered on modeling and performance evaluation, less so on data collection. However, recent research demonstrated that limitations and biases in data may negatively impact trustworthiness and reliability. These aspects are particularly impactful on sensitive domains such as mental health and neurological disorders, where speech data are used to develop AI applications for patients and healthcare providers. In this paper, we chart the landscape of available speech datasets for this domain, to highlight possible pitfalls and opportunities for improvement and promote fairness and diversity. We present a comprehensive list of desiderata for building speech datasets for mental health and neurological disorders and distill it into an actionable checklist focused on ethical concerns to foster more responsible research.

Citation

Eleonora Mancini, Ana Tanevska, Andrea Galassi, Alessio Galatolo, Federico Ruggeri, and Paolo Torroni. Promoting the responsible development of speech datasets for mental health and neurological disorders research. J. Artif. Intell. Res., 82:937–972, 2025.

@article{mancini-etal-2025-promoting-datasets,
author       = {Eleonora Mancini and
                  Ana Tanevska and
                  Andrea Galassi and
                  Alessio Galatolo and
                  Federico Ruggeri and
                  Paolo Torroni},
  title        = {Promoting the Responsible Development of Speech Datasets for Mental
                  Health and Neurological Disorders Research},
  journal      = {J. Artif. Intell. Res.},
  volume       = {82},
  pages        = {937--972},
  year         = {2025},
  url          = {https://doi.org/10.1613/jair.1.16406},
  doi          = {10.1613/JAIR.1.16406}
}