Domain Adversarial Training for Mitigating Gender Bias in Speech-based Mental Health Detection
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916722347343872 |
|---|---|
| author | Kim, June-Woo Yoon, Haram Oh, Wonkyo Jung, Dawoon Yoon, Sung-Hoon Kim, Dae-Jin Lee, Dong-Ho Lee, Sang-Yeol Yang, Chan-Mo |
| author_facet | Kim, June-Woo Yoon, Haram Oh, Wonkyo Jung, Dawoon Yoon, Sung-Hoon Kim, Dae-Jin Lee, Dong-Ho Lee, Sang-Yeol Yang, Chan-Mo |
| contents | Speech-based AI models are emerging as powerful tools for detecting depression and the presence of Post-traumatic stress disorder (PTSD), offering a non-invasive and cost-effective way to assess mental health. However, these models often struggle with gender bias, which can lead to unfair and inaccurate predictions. In this study, our study addresses this issue by introducing a domain adversarial training approach that explicitly considers gender differences in speech-based depression and PTSD detection. Specifically, we treat different genders as distinct domains and integrate this information into a pretrained speech foundation model. We then validate its effectiveness on the E-DAIC dataset to assess its impact on performance. Experimental results show that our method notably improves detection performance, increasing the F1-score by up to 13.29 percentage points compared to the baseline. This highlights the importance of addressing demographic disparities in AI-driven mental health assessment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_03359 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Domain Adversarial Training for Mitigating Gender Bias in Speech-based Mental Health Detection Kim, June-Woo Yoon, Haram Oh, Wonkyo Jung, Dawoon Yoon, Sung-Hoon Kim, Dae-Jin Lee, Dong-Ho Lee, Sang-Yeol Yang, Chan-Mo Artificial Intelligence Speech-based AI models are emerging as powerful tools for detecting depression and the presence of Post-traumatic stress disorder (PTSD), offering a non-invasive and cost-effective way to assess mental health. However, these models often struggle with gender bias, which can lead to unfair and inaccurate predictions. In this study, our study addresses this issue by introducing a domain adversarial training approach that explicitly considers gender differences in speech-based depression and PTSD detection. Specifically, we treat different genders as distinct domains and integrate this information into a pretrained speech foundation model. We then validate its effectiveness on the E-DAIC dataset to assess its impact on performance. Experimental results show that our method notably improves detection performance, increasing the F1-score by up to 13.29 percentage points compared to the baseline. This highlights the importance of addressing demographic disparities in AI-driven mental health assessment. |
| title | Domain Adversarial Training for Mitigating Gender Bias in Speech-based Mental Health Detection |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2505.03359 |