# Listen to the Features: Voice Anonymization Driven by Content Embedding Matching over Signal Reconstruction

arXiv: https://arxiv.org/abs/2607.09767 · HTML: https://mlsanity.com/paper/2607.09767/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 22 references that are in the map.
- First posted on arXiv 7 Jul 2026.

## Builds on (most cited first, 15 of 15)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022](https://mlsanity.com/paper/2204.02152/) (2022)
- [BigVGAN: A Universal Neural Vocoder with Large-Scale Training](https://mlsanity.com/paper/2206.04658/) (2022)
- [NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models](https://mlsanity.com/paper/2403.03100/) (2024)
- [Neural Networks Fail to Learn Periodic Functions and How to Fix It](https://mlsanity.com/paper/2006.08195/) (2020)
- [The VoicePrivacy 2020 Challenge: Results and findings](https://mlsanity.com/paper/2109.00648/) (2021)
- [The VoicePrivacy 2024 Challenge Evaluation Plan](https://mlsanity.com/paper/2404.02677/) (2024)
- [Speaker anonymisation using the McAdams coefficient](https://mlsanity.com/paper/2011.01130/) (2020)
- [The Third VoicePrivacy Challenge: Preserving Emotional Expressiveness and Linguistic Content in Voice Anonymization](https://mlsanity.com/paper/2601.11846/) (2026)
- [SHEET: A Multi-purpose Open-source Speech Human Evaluation Estimation Toolkit](https://mlsanity.com/paper/2505.15061/) (2025)
- [Exploring VQ-VAE with Prosody Parameters for Speaker Anonymization](https://mlsanity.com/paper/2409.15882/) (2024)
- [IDMap: A Pseudo-Speaker Generator Framework Based on Speaker Identity Index to Vector Mapping](https://mlsanity.com/paper/2511.06246/) (2025)

## Abstract (arXiv)

The paper presents a voice anonymization model focusing on preserving content rather than producing realistic speech. It relies on content embeddings extracted from a frozen pretrained wav2vec2 encoder. These embeddings are decoded into an anonymized signal using vector quantization and a HiFi-GAN vocoder, both trained on LibriTTS without any waveform reconstruction loss or speaker embedding mapping. The training objective enforces that embeddings of the anonymized signal match those of the original one. While training, an auxiliary speaker classification branch with a gradient reversal layer is used to discard speakerspecific information. Results show that this straightforward embedding-based approach achieves very low WER (2.53) with an anonymization performance (EER 13.39) ranking within first level for VPC. Notably, emotions are partially preserved (UAR 43.91), even without a supporting training objective, while the anonymized voice is audible without reconstruction loss.