A new voice analysis technology may help save millions of unknowing individuals from needless financial loss from phone scams. The technology can detect scams by analyzing and detecting suspicious phone conversations and bank transfers. The system is the work of researchers at Japan’s Nagoya University in conjunction with Fujitsu Ltd. If the technology’s high level of accuracy during trial runs is anything to go by, the potential for this software is enormous.
The system works through what the researchers termed the “over trust” condition, referring to that point at which an individual develops trust for a scammer depending on their pitch and tone. If successful, installing the technology on mobile phones could help save millions of dollars from bank transfer scams. The Japanese researchers based their technology on the fact that whenever an individual is overwhelmed by distressing information, their pitch and tone becomes flat in the high-frequency range.
At this point, they enter the situation the researchers called over trust. In over trust, individuals often suffer a reduced ability to analyze information and are vulnerable to scamming. The researchers based their understanding on the fact that most phone scammers get lucky with pretending to be relatives in distress and calling for help. Last year, Russia's Sperbank deployed a super-secure ATM machine capable of interrogating customers withdrawing money using a sophisticated lie detection system that picks up on nervousness in the voice.