Assessing the Impact of Speaker Identity in Speech Spoofing Detection
This addresses the issue of speaker bias in spoofing detection for security applications, representing an incremental improvement.
The paper tackled the problem of speaker identity affecting spoofing detection systems by proposing a Speaker-Invariant Multi-Task framework, which reduced the average equal error rate by 17% and up to 48% for challenging attacks.
Spoofing detection systems are typically trained using diverse recordings from multiple speakers, often assuming that the resulting embeddings are independent of speaker identity. However, this assumption remains unverified. In this paper, we investigate the impact of speaker information on spoofing detection systems. We propose two approaches within our Speaker-Invariant Multi-Task framework, one that models speaker identity within the embeddings and another that removes it. SInMT integrates multi-task learning for joint speaker recognition and spoofing detection, incorporating a gradient reversal layer. Evaluated using four datasets, our speaker-invariant model reduces the average equal error rate by 17% compared to the baseline, with up to 48% reduction for the most challenging attacks (e.g., A11).