Echoes Unveiled: Identifying Synthetic Voices

The advent of deep neural networks in natural language and speech processing has created a new attack vector in the form of synthetic voices cloned from less than 30-seconds of voice sample from a human counterpart. Effectively detecting spoofing attacks is critical for any speech application that uses voice for authentication, verification, and identification. With the rapid rise of highly effective speech synthesis, it is challenging to identify synthetic voices while generalizing to novel voices, synthesizers, and channel conditions. In this talk, Daniel Pluth from Vail Systems presents a model aimed at identifying synthetic voices and demonstrates its effectiveness and generalizability. He further motivates the need for the research community to consider channel conditions when detecting voice spoofing. He demonstrates that channel conditions play an inordinate role in identifying a spoofed voice, and detection techniques that do not consider variable channel conditions will exhibit high error rates.
Daniel is a Principal Data Scientist at Vail Systems. Dan's interests largely lie in spoken language understanding and have included work on speaker recognition, spoof detection and speech generation. He holds a PhD in Physics from Iowa State University, bringing a unique perspective and analytical toolkit to problem solving.