A new technology to recognize voices as a substitute for the human ear is being developed by Russian scientists

The Internet has always been filled with misinformation, but at least it was not difficult to separate facts from fiction with a little effort and focus. The emergence of advanced artificial intelligence tools has changed some facts about what can be believed or known as false, which made suspicion in everything presented or Shares on the Internet are more important than ever, as scientists of the Russian Tomsk University of Control Systems and Radio electronics have created a mathematical model that recognizes speech and outperforms its analogues with specific indicators. The innovators indicate that this model can become the basis for a comprehensive personal identification system, as well as facilitate the control of “smart” technology with the help of voice, according to a report published by the Russian “Novosti” agency.

According to experts, the method of personal identification through speech recognition is developing using digital security technology used to control equipment and security systems in remote banking services. Tomsk University of Control Systems and Radio electronics scientists have created a new mathematical model for speech recognition, which outperforms its counterparts in accuracy and allows the speaker to be identified correctly. Anton Konev, deputy director of the National Institute of Communication’s Center for Competence for Trusted Interaction Technologies, says, “The mathematical device differs from its counterparts in that it closely resembles the human inner ear when recognizing critical speech frequencies. The potential mechanisms upon which current speech recognition systems are built into this device only play a supporting role.”

Experts point out that the error rate in the new device’s recognition of the basic tone of speech is four times less than the best of its counterparts in the world, and this allows for better recognition of normal speech and even singing. It also allows communication with smart devices, not through commands, but directly, as is the case with the interlocutor. Experts assure that the new system will allow verification of the identity of the interlocutor during the communication session, which will completely eliminate some types of fraud

Leave a Reply

Your email address will not be published. Required fields are marked *