# MUSA: Multi-lingual Speaker Anonymization via Serial Disentanglement

arXiv: https://arxiv.org/abs/2407.11629 · HTML: https://mlsanity.com/paper/2407.11629/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Jul 2024.

## Built on by (most cited first, 10 of 10)

- [The Third VoicePrivacy Challenge: Preserving Emotional Expressiveness and Linguistic Content in Voice Anonymization](https://mlsanity.com/paper/2601.11846/) (2026)
- [Analysis of Speech Temporal Dynamics in the Context of Speaker Verification and Voice Anonymization](https://mlsanity.com/paper/2412.17164/) (2024)
- [Zero-Shot Voice Conversion via Content-Aware Timbre Ensemble and Conditional Flow Matching](https://mlsanity.com/paper/2411.02026/) (2024)
- [Exploiting Context-dependent Duration Features for Voice Anonymization Attack Systems](https://mlsanity.com/paper/2507.15214/) (2025)
- [Takin-VC: Expressive Zero-Shot Voice Conversion via Adaptive Hybrid Content Encoding and Enhanced Timbre Modeling](https://mlsanity.com/paper/2410.01350/) (2024)
- [SEF-MK: Speaker-Embedding-Free Voice Anonymization through Multi-k-means Quantization](https://mlsanity.com/paper/2508.07086/) (2025)
- [SegReConcat: A Data Augmentation Method for Voice Anonymization Attack](https://mlsanity.com/paper/2508.18907/) (2025)
- [Target speaker anonymization in multi-speaker recordings](https://mlsanity.com/paper/2510.09307/) (2025)
- [Anonymization, Not Elimination: Utility-Preserved Speech Anonymization](https://mlsanity.com/paper/2604.17000/) (2026)
- [Voice Privacy from an Attribute-based Perspective](https://mlsanity.com/paper/2603.20301/) (2026)

## Abstract (arXiv)

Speaker anonymization is an effective privacy protection solution designed to conceal the speaker's identity while preserving the linguistic content and para-linguistic information of the original speech. While most prior studies focus solely on a single language, an ideal speaker anonymization system should be capable of handling multiple languages. This paper proposes MUSA, a Multi-lingual Speaker Anonymization approach that employs a serial disentanglement strategy to perform a step-by-step disentanglement from a global time-invariant representation to a temporal time-variant representation. By utilizing semantic distillation and self-supervised speaker distillation, the serial disentanglement strategy can avoid strong inductive biases and exhibit superior generalization performance across different languages. Meanwhile, we propose a straightforward anonymization strategy that employs empty embedding with zero values to simulate the speaker identity concealment process, eliminating the need for conversion to a pseudo-speaker identity and thereby reducing the complexity of speaker anonymization process. Experimental results on VoicePrivacy official datasets and multi-lingual datasets demonstrate that MUSA can effectively protect speaker privacy while preserving linguistic content and para-linguistic information.