Espy-Wilson to Ensure Hearing is Believing
Professor Carol Espy-Wilson (ECE/ISR) is the principal investigator for a three-year, $500,000 National Science Foundation grant, "Nonintrusive Digital Speech Forensics: Source Identification and Content Authentication."
With the advent of the digital era, virtually every speech communication system acquires, creates, transmits, stores, and processes information in digital form. Moreover, current digital media editing software allows malicious amateurs to perform imperceptible alterations to digital content. This creates a serious threat to the "knowledge life cycle." When hearing is no longer believing, the process of going from data to information, knowledge, understanding and, finally, to decision or action is severely compromised.
To help reduce this threat, Espy-Wilson will develop theories, methods and tools for extracting and visualizing evidence from digital speech content to identify the media source and authenticate content.
Her research will be based on the important paradigm of nonintrusive media forensics. Unlike digital watermarking techniques, the approach assumes no explicit prior knowledge about the original speech signal. The underlying hypothesis of the methodology is that physical devices, along with their associated signal processing chain, leave behind intrinsic fingerprint traces in the speech signal that are detectable by statistical methods.
September 18, 2009