Russian Researchers Develop Voice Authentication System That Detects Fatigue and Alcohol Intoxication
Scientists in Russia have created a voice authentication technology that can identify both a speaker and their physical condition.
Researchers at Omskiy gosudarstvennyy tekhnicheskiy universitet (Omsk State Technical University) have developed a voice authentication method that can also analyze a user’s condition. The technology can distinguish between a normal state, drowsiness, and several stages of alcohol intoxication.
The discovery was registered by Daniil Inivatov, an assistant at the Kompleksnaya zashchita informatsii (Comprehensive Information Security) department at OmGTU. Russia’s Ministry of Science and Higher Education reported the development.
The new technology could be used to protect sensitive data in biometric access systems and to prevent user errors caused by altered states of consciousness.
Security Technology
Voice biometrics are already used in banking, call centers, and smart devices. However, standard systems verify only the identity of the speaker.
The new system analyzes speech using additional parameters. A neural network evaluates pitch, timbre, overtones, pauses between words, and speech rhythm. These characteristics change when a person is fatigued, stressed, or under the influence of alcohol.
According to developer Daniil Inivatov, allowing a user to access data or make decisions in such a condition can create risks both for information systems and for the individual.
Limiting Access
If the algorithm detects severe fatigue or signs of intoxication, the system can restrict a user’s actions. In certain cases, it may temporarily block access to specific operations.
The mechanism could be applied in financial systems, corporate services, and infrastructure with heightened security requirements.
Developers believe the technology could reduce risks associated with human error when working with critical data and increase the overall reliability of digital authentication systems.
