Challenged the previous disentanglement assumption in feature extraction process, by removing residual speaker information from speaker-independent attributes (linguistic and prosodic features). The removal is achieved by adding differentially-private noise in these features, which allows us to provide formal provable guarantees of privacy leakage.

We investigate the effect of various design choices in x-vector based speaker anonymization method, on Privacy and Utility. Some choices seem to be more robust than others in VoicePrivacy challenge setup.

We aim a paradigm shift in context of speaker privacy evaluation from “security by obscurity” to Semi-Informed and Informed attackers. We show that privacy obtained by voice transformation techniques can be breached by an informed attacker.

In this work, we propose a privacy-preserving framework based on speaker-adversarial training of end-to-end ASR. We evaluate the system using closed-set and open-set identification and observe a strange disparity in results.

In this work, we propose a large scale spoken language identification technique over 176 languages. We also provide evidence that the languages are clustered based on their geographical and ethnological proximity.