Shulou(Shulou.com)11/24 Report--

CTOnews.com, June 29 (Xinhua)-- computer scientists at the University of Waterloo recently developed a new voice Deepfake software, which has a 99% chance of successfully deceiving the voice authentication system.

Image source: uk.pcmag.com this software uses machine learning software and only needs 5 minutes of human voice recording to simulate a very realistic human voice.

Andre Kassis, Ph.D., computer Security and Privacy, University of Waterloo, explained how voice authentication works. CTOnews.com translated as follows:

After the user registers through voice authentication, the user will be asked to repeat a specific phrase or sentence.

The system extracts the voice print (voice fingerprint) based on the user's voice and stores it on the server.

If you try to authenticate in the future, you will be prompted to say a different phrase, and the features extracted from it will be compared with the voice fingerprint stored in the system to determine whether access should be granted.

The team led by Kassis found that there were obvious computer-generated features in existing deepfake software and developed a program to eliminate these traces and make it indistinguishable from real audio.

In the team's test of the Amazon Connect voice authentication system, the pass rate was 10% within 4 seconds and 40% within 30 seconds. After 6 consecutive attempts, the team thought the success rate could reach 99%.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.