MADISON — Researchers are in an arms race with hackers to prevent data theft. Their standard tools include strategies like multi-factor authentication systems, fingerprint technology and retinal scans. One type of security system that is gaining popularity is automatic speaker identification, which uses a person’s voice as a passcode.
These systems, already in use for phone banking and other applications, are good at weeding out attacks that try to fake a user’s voice through digital manipulation. But digital security engineers at the University of Wisconsin–Madison have found these systems are not quite as foolproof when it comes to a novel analog attack. They found that speaking through customized PVC pipes — the type found at most hardware stores — can trick machine learning algorithms that support automatic speaker identification systems.
The team, led by PhD…
