Using Deep Learning for Detecting Spoofing Attacks on Speech Signals
This work addresses security vulnerabilities in speaker verification systems, but it is incremental as it applies existing deep learning methods to a standard challenge dataset.
The paper tackled the problem of spoofing attacks on speaker verification systems by developing deep learning-based systems for the ASVSpoof2015 Challenge, achieving less than 0.5% Equal Error Rate for known attacks.
It is well known that speaker verification systems are subject to spoofing attacks. The Automatic Speaker Verification Spoofing and Countermeasures Challenge -- ASVSpoof2015 -- provides a standard spoofing database, containing attacks based on synthetic speech, along with a protocol for experiments. This paper describes CPqD's systems submitted to the ASVSpoof2015 Challenge, based on deep neural networks, working both as a classifier and as a feature extraction module for a GMM and a SVM classifier. Results show the validity of this approach, achieving less than 0.5\% EER for known attacks.