Exploring WavLM Back-ends for Speech Spoofing and Deepfake Detection
This work addresses the problem of detecting spoofed or deepfake speech for security applications, but it is incremental as it builds on existing self-supervised models and challenge frameworks.
The paper tackled speech deepfake detection by leveraging a pre-trained WavLM model with various back-end techniques and data augmentations, achieving results such as 3.42% EER and 0.0937 minDCF in the ASVspoof 5 Challenge.
This paper describes our submitted systems to the ASVspoof 5 Challenge Track 1: Speech Deepfake Detection - Open Condition, which consists of a stand-alone speech deepfake (bonafide vs spoof) detection task. Recently, large-scale self-supervised models become a standard in Automatic Speech Recognition (ASR) and other speech processing tasks. Thus, we leverage a pre-trained WavLM as a front-end model and pool its representations with different back-end techniques. The complete framework is fine-tuned using only the trained dataset of the challenge, similar to the close condition. Besides, we adopt data-augmentation by adding noise and reverberation using MUSAN noise and RIR datasets. We also experiment with codec augmentations to increase the performance of our method. Ultimately, we use the Bosaris toolkit for score calibration and system fusion to get better Cllr scores. Our fused system achieves 0.0937 minDCF, 3.42% EER, 0.1927 Cllr, and 0.1375 actDCF.