Why disentanglement-based speaker anonymization systems fail at preserving emotions?
Gaznepoglu ÜE, Peters N (2025)
Publication Type: Conference contribution
Publication year: 2025
Journal
Publisher: Institute of Electrical and Electronics Engineers Inc.
Conference Proceedings Title: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Event location: Hyderabad, IND
DOI: 10.1109/ICASSP49660.2025.10889709
Abstract
Disentanglement-based speaker anonymization involves decomposing speech into a semantically meaningful representation, altering the speaker embedding, and resynthesizing a waveform using a neural vocoder. State-of-the-art systems of this kind are known to remove emotion information. Possible reasons include mode collapse in GANbased vocoders, unintended modeling and modification of emotions through speaker embeddings, or excessive sanitization of the intermediate representation (IR). In this paper, we conduct a comprehensive evaluation of a state-of-the-art speaker anonymization system to understand the underlying causes. We conclude that the main reason is the lack of emotion-related information in the IR. The speaker embeddings also have a high impact, if they are learned in a generative context. The vocoder's out-of-distribution performance has a smaller impact. Additionally, we discovered that synthesis artifacts increase spectral kurtosis, biasing emotion recognition evaluation towards classifying utterances as angry. Therefore, we conclude that reporting unweighted average recall alone for emotion recognition performance is suboptimal.
Authors with CRIS profile
Involved external institutions
Ireland (IE)How to cite
APA:
Gaznepoglu, Ü.E., & Peters, N. (2025). Why disentanglement-based speaker anonymization systems fail at preserving emotions? In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Hyderabad, IND: Institute of Electrical and Electronics Engineers Inc..
MLA:
Gaznepoglu, Ünal Ege, and Nils Peters. "Why disentanglement-based speaker anonymization systems fail at preserving emotions?" Proceedings of the 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025, Hyderabad, IND Institute of Electrical and Electronics Engineers Inc., 2025.
BibTeX: Download