Attentive Merging of Hidden Embeddings from Pre-trained Speech Model for Anti-spoofing Detection
Open accessThe DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.
Abstract
Self-supervised learning (SSL) speech representation models, trained on large speech corpora, have demonstrated effectiveness in extracting hierarchical speech embeddings through multiple transformer layers. However, the behavior of these embeddings in specific tasks remains uncertain. This paper investigates the multi-layer behavior of the WavLM model in anti-spoofing and proposes an attentive merging method to leverage the hierarchical hidden embeddings. Results demonstrate the feasibility of fine-tuning WavLM to achieve the best equal error rate (EER) of 0.65%, 3.50%, and 3.19% on the ASVspoof 2019LA, 2021LA, and 2021DF evaluation sets, respectively. Notably, We find that the early hidden transformer layers of the WavLM large model contribute significantly to anti-spoofing task, enabling computational efficiency by utilizing a partial pre-trained model.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Cited by
Mixture of Experts Fusion for Fake Audio Detection Using Frozen wav2vec 2.0
Wang, Z., Fu, R., Wen, Z., Tao, J., Wang, X., Xie, Y., Qi, X., Shi, S., Lu, Y., Liu, Y., et al. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10888990 · Cited by 54
Generalizable Audio Deepfake Detection via Latent Space Refinement and Augmentation
Huang, W. C., Gu, Y., Wang, Z., Zhu, H., & Qian, Y. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10888328 · Cited by 45
Nes2Net: A Lightweight Nested Architecture for Foundation Model Driven Speech Anti-Spoofing
Liu, T., Truong, D.-T., Das, R. K., Lee, K. A., & Li, H. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3626963 · Cited by 69
Self-citationGuragain, A., Liu, T., Pan, Z., Sailor, H. B., & Wang, Q. (2024) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt61566.2024.10832226 · Cited by 39
Self-citationSVDD 2024: The Inaugural Singing Voice Deepfake Detection Challenge
Zhang, Y., Zang, Y., Shi, J., Yamamoto, R., Toda, T., & Duan, Z. (2024) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt61566.2024.10832284 · Cited by 20
Metadata
- Authors
- Pan, Z., Liu, T., Sailor, H. B., & Wang, Q.
- Venue
- Interspeech 2024
- Publisher
- ISCA
- Published
- 2024-09-01
- Pages
- 2090-2094
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W4402111788
- Cited by
- 32
- References
- 0
- Citations by year
- 2026: 13 · 2025: 14 · 2024: 5
- FWCI
- 9.44
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Interspeech (USB) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Research Foundation
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Signal Processing › Biometric Identification and Security
- Keywords
- voice spoofing, pre-trained speech models, attention mechanism, speaker verification security
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.