VoxAnchor: Grounding Speech Authenticity in Throat Vibration via mmWave Radar
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917365954904064 |
|---|---|
| author | Han, Mingda Yang, Huanqi Li, Chaoqun Li, Wenhao Zhang, Guoming Yang, Yanni Cao, Yetong Xu, Weitao Hu, Pengfei |
| author_facet | Han, Mingda Yang, Huanqi Li, Chaoqun Li, Wenhao Zhang, Guoming Yang, Yanni Cao, Yetong Xu, Weitao Hu, Pengfei |
| contents | Rapid advances in speech synthesis and audio editing have made realistic forgeries increasingly accessible, yet existing detection methods remain vulnerable to tampering or depend on visual/wearable sensors. In this paper, we present VoxAnchor, a system that physically grounds audio authentication in vocal dynamics by leveraging the inherent coherence between speech acoustics and radar-sensed throat vibrations. VoxAnchor uses contactless millimeter-wave radar to capture fine-grained throat vibrations that are tightly coupled with human speech production, establishing a hard-to-forge anchor rooted in human physiology. The design comprises three main components: (1) a cross-modal frame-work that uses modality-specific encoders and contrastive learning to detect subtle mismatches at word granularity; (2) a phase-aware pipeline that extracts physically consistent, temporally faithful throat vibrations; and (3) a dual-stage strategy that combines signal-level onset detection and semantic-level coherence to align asynchronous radar and audio streams. Unlike liveness detection, which only confirms whether speech occurred, VoxAnchor verifies what was spoken through word-level content consistency, exposing localized edits that preserve identity and global authenticity cues. Extensive evaluations show that VoxAnchor achieves robust, fine-grained detection across diverse forgeries (editing, splicing, replay, deepfake) and conditions, with an overall EER of 0.017, low latency, and modest computational cost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_27562 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | VoxAnchor: Grounding Speech Authenticity in Throat Vibration via mmWave Radar Han, Mingda Yang, Huanqi Li, Chaoqun Li, Wenhao Zhang, Guoming Yang, Yanni Cao, Yetong Xu, Weitao Hu, Pengfei Human-Computer Interaction Rapid advances in speech synthesis and audio editing have made realistic forgeries increasingly accessible, yet existing detection methods remain vulnerable to tampering or depend on visual/wearable sensors. In this paper, we present VoxAnchor, a system that physically grounds audio authentication in vocal dynamics by leveraging the inherent coherence between speech acoustics and radar-sensed throat vibrations. VoxAnchor uses contactless millimeter-wave radar to capture fine-grained throat vibrations that are tightly coupled with human speech production, establishing a hard-to-forge anchor rooted in human physiology. The design comprises three main components: (1) a cross-modal frame-work that uses modality-specific encoders and contrastive learning to detect subtle mismatches at word granularity; (2) a phase-aware pipeline that extracts physically consistent, temporally faithful throat vibrations; and (3) a dual-stage strategy that combines signal-level onset detection and semantic-level coherence to align asynchronous radar and audio streams. Unlike liveness detection, which only confirms whether speech occurred, VoxAnchor verifies what was spoken through word-level content consistency, exposing localized edits that preserve identity and global authenticity cues. Extensive evaluations show that VoxAnchor achieves robust, fine-grained detection across diverse forgeries (editing, splicing, replay, deepfake) and conditions, with an overall EER of 0.017, low latency, and modest computational cost. |
| title | VoxAnchor: Grounding Speech Authenticity in Throat Vibration via mmWave Radar |
| topic | Human-Computer Interaction |
| url | https://arxiv.org/abs/2603.27562 |