Studies show that Voice impersonation and replays using advanced audio tools can trick the system to breach authentication. Even though Voice biometrics are improving but the question arises in their reliability.
Voice biometrics is already being in use, for instance, in Google Home. It allows users Google Assistant to seek users’ personal information after it recognizes their voice. Also in banks, like HSBC is offering a voice biometric system that, while creating a voice ID, looks for 100 identifiers such as accent, pronunciation, vocal tract, the role of the tongue and nasal cavity, to verify log-in by a customer.
Persistent Systems and ValidSoft have jointly developed and designed a secure digital voice authentication. Biometric will first process the input voice to extract its characteristics that are specific to a speaker to build voice print or voice signature. To verify the new input, the same process is repeated and a similarity measure is calculated.
Also, it is designed to deal with variability like if a user has a cold and sounds different but to an advanced biometric engine it will sound just the same.
Researchers at HSE University and Nizhny Novgorod State Linguistic University have developed an artificial intelligence (AI) based voice recognition system, which can reduce the error rate of such systems to 2% at a signal-to-noise ratio of 10dB or higher.
Harnessing the power of AI engines, voice biometrics and natural language understanding (NLU) could be leveraged to authenticate individuals to a much higher degree of accuracy, making voice biometrics also a reasonably viable fingerprint of the future.
Venkat Krishnapur, vice president of engineering and managing director at McAfee India
Advanced recognition engines can also detect replay attacks by looking for the highest and lowest frequencies or by detecting distortion caused by the replaying of audio. Though the tech has evolved, a lot more needs to be done to not just improve security but also its efficiency.