Avaliações PREreview de “Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models”

Ir para detalhes do preprint Ir para avaliações PREreview

Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models

de Nikita Kuzmin, Tao Zhong, Jiajun Deng, Yingke Zhu, Tristan Tsoi, Tianxiang Cao, Simon Lui, Kong Aik Lee e Eng Siong Chng

Publicado: 9 de março de 2026
Servidor: arXiv
DOI: 10.48550/arxiv.2603.08179

Resumo

End-to-end full-duplex speech models feed user audio through an always-on LLM backbone, yet the speaker privacy implications of their hidden representations remain unexamined. Following the VoicePrivacy 2024 protocol with a lazy-informed attacker, we show that the hidden states of SALM-Duplex and Moshi leak substantial speaker identity across all transformer layers. Layer-wise and turn-wise analyses reveal that leakage persists across all layers, with SALM-Duplex showing stronger leakage in early layers while Moshi leaks uniformly, and that Linkability rises sharply within the first few turns. We propose two streaming anonymization setups using Stream-Voice-Anon: a waveform-level front-end (Anon-W2W) and a feature-domain replacement (Anon-W2F). Anon-W2F raises EER by over 3.5x relative to the discrete encoder baseline (11.2% to 41.0%), approaching the 50% random-chance ceiling, while Anon-W2W retains 78-93% of baseline sBERT across setups with sub-second response latency (FRL under 0.8 s).

Ler o preprint

1 Avaliação PREreview

Escrever uma Avaliação PREreview Solicitar uma Avaliação PREreview

Avaliação PREreview de shireesh apte

De autoria de shireesh apte

summary: 'Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models by Nikita Kuzmin, Tao Zhong, Jiajun Deng (corresponding author), Yingke Zhu, Tristan Tsoi, Tianxiang Cao, Simon Lui, Kong Aik Lee, and Eng Siong Chng audits speaker identity leakage in hidden states of always-on,…

Ler a avaliação PREreview de shireesh apte

Avaliações PREreview de Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models

1 Avaliação PREreview