Latin America. As people become increasingly comfortable with biometrics, voice authentication is finding more applications across industries, such as healthcare, banking, and education.
The voice biometrics market will grow at a compound annual growth rate (CAGR) of 19.4% between 2017 and 2021. Voice recognition systems monitor tone of voice and accent, as well as indicate the shape and size of the larynx, nasal passage, and vocal tract to help identify and authenticate a person.
How does the system work? The user must first create a voice impression by recording their name, age, address, and a series of secret sound notes. The speech recognition system captures the voice impression and analyzes the pattern of expression and breathing. The speech print is then encrypted and stored in Active Directory as part of the user's authentication profile, along with other authentication credentials. Each time a user with voice authentication calls over the phone, their voice is compared to the print stored in Active Directory, which speeds up and optimizes authentication.
How companies are using voice biometrics for authentication
Voice authentication offers a flexible and cost-effective form of biometric authentication, as it does not require the hardware integration that might be needed for other methods, such as fingerprint correlation or retina reader. Businesses can use voice recognition to improve security and give customers a choice as to how they want to authenticate. Mentioned below are some use cases of voice-based authentication in different industries:
Entertainment: Voice recognition can be used to change TV or radio channels, open and close screens, and play movies. It can also help personalize the customer experience. For example, services like Netflix and Hulu can be customized to determine your age using voice analytics so that they have access to age-appropriate content.
Bless you: The global medical biometrics market is anticipated to reach $14.5 billion by 2025, according to a recent report by Grand View Research, Inc. In an industry where data protection is paramount, doctors can use voice biometrics to dictate and record the patient's health status directly in the system and consult their medical history securely. This can significantly benefit patients who need to share their record with multiple doctors. The system can also help significantly reduce fraud to providers and customers by automating collections and improving patient satisfaction by offering them different payment options.
Banking: Customers can use voice authentication to operate their safe deposit boxes at banks. Banks, on the other hand, can use it to implement a highly secure and advanced voice-based payment system. In light of the rise in fraud, credit card issuers and banks such as Citibank and ANZ are using voice biometrics to proactively identify fraudsters and authenticate callers to their call centers.
Education: Educational institutions can use voice recognition to offer flexibility to visually impaired students; for example, helping them take online exams through voice authentication.
Independent Software Vendors (ISVs): For ISVs, voice authentication enables enterprises to apply login mechanisms similar to those based on Active Directory to unify authentication across enterprise applications and strengthen accessibility compliance.
The Challenges of Voice-Based Authentication
While each person has a unique voice, secure authentication through voice recognition can be challenging in certain cases; for example, when the user has the flu or is hoarse. Therefore, it is important to prevent unauthorized users from accessing the database by mimicking someone else's voice.
The best way to avoid this is to approve speech prints and store them in Active Directory, a process in which the customer using the speech recognition system registers with an authorized member database and their voice printing is used as valid authentication.
With Active Directory, each registered member's unique tone of voice is compared to both a whitelist of valid voice impressions and a blacklist of fraudulent voice impressions. While the whitelist authentication process is taking place, passive fraud detection can be enabled to issue a real-time alert—in case the caller's voice correlates with a record in the blacklist database.
