EnvTriCascade: An Environment-Aware Tri-Stage Cascaded Framework for ESDD2 2026 Challenge
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911694271283200 |
|---|---|
| author | Huang, Hengyan Guo, Xiaoxuan Zhou, Jiayi Xie, Yuankun Liu, Jian Cheng, Haonan Ye, Long Zhang, Qin |
| author_facet | Huang, Hengyan Guo, Xiaoxuan Zhou, Jiayi Xie, Yuankun Liu, Jian Cheng, Haonan Ye, Long Zhang, Qin |
| contents | ADD in real-world scenarios has evolved from speech-only spoofing to more challenging component-level settings, where speech and environmental sounds may be independently manipulated. To tackle this, we propose EnvTriCascade, an Environment-Aware Tri-Stage Cascaded framework for the ESDD2 Challenge. First, a mix-consistency detector provides a binary prior to distinguish original recordings from manipulated mixtures, which calibrates the final decisions. Next, two complementary five-class detectors, leveraging SSLAM+XLS-R and EAT-large+XLS-R representations, extract robust multi-branch features integrated via a cross-branch attention-gated classifier. To enhance robustness against diverse mixing conditions, we incorporate RawBoost augmentation. Trained exclusively on the official CompSpoofV2 dataset, our system achieves a Macro-F1 score of 0.8266 on the test set, significantly outperforming the official baseline and ranking second in the challenge. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_18409 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EnvTriCascade: An Environment-Aware Tri-Stage Cascaded Framework for ESDD2 2026 Challenge Huang, Hengyan Guo, Xiaoxuan Zhou, Jiayi Xie, Yuankun Liu, Jian Cheng, Haonan Ye, Long Zhang, Qin Sound ADD in real-world scenarios has evolved from speech-only spoofing to more challenging component-level settings, where speech and environmental sounds may be independently manipulated. To tackle this, we propose EnvTriCascade, an Environment-Aware Tri-Stage Cascaded framework for the ESDD2 Challenge. First, a mix-consistency detector provides a binary prior to distinguish original recordings from manipulated mixtures, which calibrates the final decisions. Next, two complementary five-class detectors, leveraging SSLAM+XLS-R and EAT-large+XLS-R representations, extract robust multi-branch features integrated via a cross-branch attention-gated classifier. To enhance robustness against diverse mixing conditions, we incorporate RawBoost augmentation. Trained exclusively on the official CompSpoofV2 dataset, our system achieves a Macro-F1 score of 0.8266 on the test set, significantly outperforming the official baseline and ranking second in the challenge. |
| title | EnvTriCascade: An Environment-Aware Tri-Stage Cascaded Framework for ESDD2 2026 Challenge |
| topic | Sound |
| url | https://arxiv.org/abs/2605.18409 |