Introduces DSFA to turn deterministic audio features stochastic during fine-tuning and the CoSG ExtEval dataset, claiming SOTA generalization for CodecFake detection.
Beyond the benchmarks (a)–(f) from CodecFake+ [10] on existing sets (ASVspoof19 LA, CoRS, CoSG Eval), we further evaluate per- formance on our new collected CoSG ExtEval dataset
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Mitigating Proxy-to-Wild Domain Gap in Deepfake Speech
Introduces DSFA to turn deterministic audio features stochastic during fine-tuning and the CoSG ExtEval dataset, claiming SOTA generalization for CodecFake detection.