“Is it true?" Audio recognition and tampering detection as means for authenticating communications

Abstract

Today, many people are concerned about the privacy of their conversations. But a related, even bigger problem looms: If communications can be intercepted, they can also be modified. How can I trust even spoken communications, have recordings been changed to modify their meaning? The talk will introduce some newer technologies to address these questions: We can identify melodies or even one certain recording of music. Using similar technologies, and also completely different approaches based on signal analysis, modifications in audio material including spoken words can be identified. This can be used to detect whether e.g. the controversial recording of some politician has been tampered with.