analiza aiqbench
This research advances spoofed speech detection by introducing a Temporal Pyramid Adapter that leverages multi-scale temporal convolutions to identify artifacts across local and global prosodic irregularities. By integrating self-supervised XLS-R representations, the model achieves superior cross-dataset generalization on benchmarks like ASVspoof 2017/2021 while highlighting specific challenges in adapting to domain or language shifts.
Robust Spoofed Speech Detection via Temporal Pyramid Modeling
arXiv:2606.16837v1 Announce Type: cross Abstract: Spoofed speech detection is increasingly challenged by realistic synthesis, voice conversion, and replay attacks, with cross-dataset generalization remaining a major limitation. This work we propose a Temporal Pyramid Adapter that utilize parallel temporal convolutions with varying receptive fields to capture multi-scale spoofing cues, ranging from local artifacts to global prosodic irregularities. We also integrated self-supervised XLS-R representations combined with front-end adapters, including Mel, Sinc, and a Temporal Pyramid design for multi-scale temporal modeling. The proposed model is evaluated cross multiple benchmark including ASVspoof 2017, ASVspoof 2021 (DF/LA), PartialSpoof, DiffSSD, and multilingual HQ-MPSD datasets. Experimental results demonstrate that Temporal Pyramid model obtained AUC of 99.24% and a EER of 3.87% on the PartialSpoof database, which is significantly outperforming the base model and several SOTA baseline such as LCNN-BLSTM (9.87% EER) and TRACE (8.08% EER). Additionally, multilingual evaluations confirm that while spoofing artifact are independent from language. While self-supervised representations improve robustness, performance degrades under domain and language shifts, highlighting the need for better adaptation and calibration strategies.
Abstract:Spoofed speech detection is increasingly challenged by realistic synthesis, voice conversion, and replay attacks, with cross-dataset generalization remaining a major limitation. This work we propose a Temporal Pyramid Adapter that utilize parallel temporal convolutions with varying receptive fields to capture multi-scale spoofing cues, ranging from local artifacts to global prosodic irregularities. We also integrated self-supervised XLS-R representations combined with front-end adapters, including Mel, Sinc, and a Temporal Pyramid design for multi-scale temporal modeling. The proposed model is evaluated cross multiple benchmark including ASVspoof 2017, ASVspoof 2021 (DF/LA), PartialSpoof, DiffSSD, and multilingual HQ-MPSD datasets. Experimental results demonstrate that Temporal Pyramid model obtained AUC of 99.24% and a EER of 3.87% on the PartialSpoof database, which is significantly outperforming the base model and several SOTA baseline such as LCNN-BLSTM (9.87% EER) and TRACE (8.08% EER). Additionally, multilingual evaluations confirm that while spoofing artifact are independent from language. While self-supervised representations improve robustness, performance degrades under domain and language shifts, highlighting the need for better adaptation and calibration strategies.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Sound (cs.SD)
arXiv:2606.16837 [cs.CV]
(or arXiv:2606.16837v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2606.16837 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 268 cuvinte
Subiecte conexe
Time-Frequency Consistency Learning for Robust Speech Deepfake Detection
arXiv · 21 iul. 2026
ArFake: A Robust Framework for Multi-Dialect Arabic Speech Spoofing Detection Benchmark
arXiv · 17 iun. 2026
Trajectory Dynamics in Self-Supervised Learning Latent Space for Audio Deepfake Detection
arXiv · 17 aug. 2026
From Self-Supervised Speech Models to Mixture-of-Experts for Robust Anti-Spoofing
arXiv · 15 iun. 2026