Machine Unlearning in Speech Emotion Recognition via Forget Set Alone
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912781528203264 |
|---|---|
| author | Ren, Zhao Rammohan, Rathi Adarshi Scheck, Kevin Schultz, Tanja |
| author_facet | Ren, Zhao Rammohan, Rathi Adarshi Scheck, Kevin Schultz, Tanja |
| contents | Speech emotion recognition aims to identify emotional states from speech signals and has been widely applied in human-computer interaction, education, healthcare, and many other fields. However, since speech data contain rich sensitive information, partial data can be required to be deleted by speakers due to privacy concerns. Current machine unlearning approaches largely depend on data beyond the samples to be forgotten. However, this reliance poses challenges when data redistribution is restricted and demands substantial computational resources in the context of big data. We propose a novel adversarial-attack-based approach that fine-tunes a pre-trained speech emotion recognition model using only the data to be forgotten. The experimental results demonstrate that the proposed approach can effectively remove the knowledge of the data to be forgotten from the model, while preserving high model performance on the test set for emotion recognition. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_04251 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Machine Unlearning in Speech Emotion Recognition via Forget Set Alone Ren, Zhao Rammohan, Rathi Adarshi Scheck, Kevin Schultz, Tanja Sound Audio and Speech Processing Speech emotion recognition aims to identify emotional states from speech signals and has been widely applied in human-computer interaction, education, healthcare, and many other fields. However, since speech data contain rich sensitive information, partial data can be required to be deleted by speakers due to privacy concerns. Current machine unlearning approaches largely depend on data beyond the samples to be forgotten. However, this reliance poses challenges when data redistribution is restricted and demands substantial computational resources in the context of big data. We propose a novel adversarial-attack-based approach that fine-tunes a pre-trained speech emotion recognition model using only the data to be forgotten. The experimental results demonstrate that the proposed approach can effectively remove the knowledge of the data to be forgotten from the model, while preserving high model performance on the test set for emotion recognition. |
| title | Machine Unlearning in Speech Emotion Recognition via Forget Set Alone |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2510.04251 |