Cyara Solutions:
Voice Biometrics Testing
Automated Testing for Enrollment and Verification
Voice biometrics adoption is experiencing strong growth, driven by the need for foolproof security methods for technology-driven systems and the popularity of voice assistants such as, Alexa, Google Assistant, and Siri. The growth rate for the global voice recognition biometrics market CAGR is 19.6% for 2017-20211. The face and voice biometrics market is expected to reach $24 billion by 20252.
Companies have had to stem the rising tide of identity fraud over the phone as fraudsters call contact centers and manipulate their way into accounts. Username and password have proven to be insufficient authentication factors, while biometrics identification has shown to be the safest for fraud prevention3. Using voice authentication is similar to using a fingerprint, as credentials cannot be altered because of the use of accurate algorithms for authentication process. Simple spoken passphrases can be recorded with speech, pitch, voice, and range of an individual used as voice recognition parameters. Voice biometrics offers a high degree of accuracy and does not require additional hardware such as iris-scanning or fingerprint sensor, aside from a microphone.
Companies and Industries Using Voice Biometrics
Banks have been deploying voice authentication software for some time with Barclays, HSBC, National Australia Bank, and others using it for phone and mobile banking. Apple’s TouchID, Clear’s airport security offering and Vodacom are the latest large deployments. Ecommerce, healthcare, and government are the other industries actively using the technology.
Voice biometrics authentication works by recording a voice and translating the record into a voiceprint. Algorithms are used to determine if the data points spoken into a phone match those in the voiceprint. Customers must have a voiceprint made either through active or passive authentication. Active authentication requires two pieces of data, a pass phrase, and some other information. A passive authentication is where a customer’s initial conversation with an agent is recorded and translated into a voiceprint. In both instances, the customer’s voice is compared to the stored voiceprint to verify the person.
The Need for Automated Testing of Voice Biometrics
Companies have the challenge of testing voice biometrics systems to ensure that it works smoothly for customers. The aspects of the systems that need to be tested are:
- Functional and regression testing of the systems enrolling and verifying customer voices
- Environmental factors testing for noise and phone equipment
- Performance testing of enrollment and verification during high volume scenarios
- Monitoring for performance and other issues
Companies need an automated solution to test the enrollment and verification of thousands of customer voiceprints. Manual testing using agents or other staff is slow and often inaccurate. In addition, manual testing cannot easily perform load testing for high volume scenarios.
The use of biometrics for authentication will continue to grow with facial recognition and multi-modal innovations on the way. It is important that companies find automated ways to thoroughly test this authentication technology for scalability and accuracy.
Learn about the Cyara Automated CX Assurance Platform and the Cyara Velocity product. Read about Voice Quality testing and Can You Hear Me Now? Best Practices in Voice Quality Monitoring.
Learn More
See how the Cyara Automated Customer Experience Platform can help you achieve your CX Assurance goals, and delight your customers.
1 “Global Voice Recognition Biometrics 2017-2021.” Technavio, Febuary 8, 2017
2 “Face and Voice Biometrics Market to Reach US $24 Bn Value by 2025.” Persistence Market Research, June 19, 2017
3 “Global Voice Biometrics Market.” Transparency, 2017