Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.23096 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915579789574144 |
|---|---|
| author | Hong, Jiyoung Chung, Yoonseo Oh, Seungyeon Kim, Juntae Lee, Jiyoung Kim, Sookyung Cho, Hyunsoo |
| author_facet | Hong, Jiyoung Chung, Yoonseo Oh, Seungyeon Kim, Juntae Lee, Jiyoung Kim, Sookyung Cho, Hyunsoo |
| contents | Audio deepfakes pose a growing threat, already exploited in fraud and misinformation. A key challenge is ensuring detectors remain robust to unseen synthesis methods and diverse speakers, since generation techniques evolve quickly. Despite strong benchmark results, current systems struggle to generalize to new conditions limiting real-world reliability. To address this, we introduce TWINSHIFT, a benchmark explicitly designed to evaluate detection robustness under strictly unseen conditions. Our benchmark is constructed from six different synthesis systems, each paired with disjoint sets of speakers, allowing for a rigorous assessment of how well detectors generalize when both the generative model and the speaker identity change. Through extensive experiments, we show that TWINSHIFT reveals important robustness gaps, uncover overlooked limitations, and provide principled guidance for developing ADD systems. The TWINSHIFT benchmark can be accessed at https://github.com/intheMeantime/TWINSHIFT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_23096 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TwinShift: Benchmarking Audio Deepfake Detection across Synthesizer and Speaker Shifts Hong, Jiyoung Chung, Yoonseo Oh, Seungyeon Kim, Juntae Lee, Jiyoung Kim, Sookyung Cho, Hyunsoo Sound Audio deepfakes pose a growing threat, already exploited in fraud and misinformation. A key challenge is ensuring detectors remain robust to unseen synthesis methods and diverse speakers, since generation techniques evolve quickly. Despite strong benchmark results, current systems struggle to generalize to new conditions limiting real-world reliability. To address this, we introduce TWINSHIFT, a benchmark explicitly designed to evaluate detection robustness under strictly unseen conditions. Our benchmark is constructed from six different synthesis systems, each paired with disjoint sets of speakers, allowing for a rigorous assessment of how well detectors generalize when both the generative model and the speaker identity change. Through extensive experiments, we show that TWINSHIFT reveals important robustness gaps, uncover overlooked limitations, and provide principled guidance for developing ADD systems. The TWINSHIFT benchmark can be accessed at https://github.com/intheMeantime/TWINSHIFT. |
| title | TwinShift: Benchmarking Audio Deepfake Detection across Synthesizer and Speaker Shifts |
| topic | Sound |
| url | https://arxiv.org/abs/2510.23096 |