Almanac
benchmark

ASVspoof 5

benchmarkactiveprovisionalasvspoof-5-baba9ac3·2 events·first seen 7d ago

Aliases: ASVspoof 5

Co-occurring entities

More like this (12)

Recent events (2)

5arXiv · cs.AI·7d ago·source ↗

RAT: Reference-Augmented Training improves deepfake audio detection without reference at inference

Researchers introduce Reference-Augmented Training (RAT), a training strategy for automatic speaker verification (ASV) anti-spoofing that conditions a model on speaker-reference recordings during training but discovers the model learns to ignore the reference at inference. Counterintuitively, this training regime induces invariances that improve deepfake detection even when the reference is replaced with a zero vector at test time. RAT achieves state-of-the-art 2.57% EER and 0.074 minDCF on the ASVspoof 5 benchmark with a single detector, outperforming large ensemble systems.

5arXiv · cs.AI·7d ago·source ↗

Explainability pipeline reveals divergent cues used by deepfake speech detectors

Researchers propose an audio-native explainability pipeline using Integrated Gradients on time-aligned self-supervised representations to localize decision evidence in deepfake speech detectors. Applied to three WavLM-based detectors (AASIST, CA-MHFA, SLS) on the ASVspoof 5 benchmark, the method reveals that despite similar performance, each detector relies on fundamentally different cues: environmental noise, phoneme artifacts, and word boundaries respectively. Findings are validated via causal masking experiments that confirm performance degrades when primary cues are removed. The work advances interpretability of audio deepfake detection, relevant to AI safety and media authenticity.