Speech Technology Magazine

 

The State of Voice Biometrics

Voice biometrics is becoming popular with big corporations but is still out of reach for many
By Phillip Britt - Posted Mar 11, 2019
« previous Page2 of 2
Bookmark and Share

Outside of financial services, another growth area for voice biometrics is expected to be healthcare, according to Grand View Research, which expects a compound annual growth rate of more than 24% for all biometrics in the industry as it tries to comply with regulations governing patient identity and patient records.

However, even though voice transactions are growing as customers—particularly Millennials and Generation Z—opt for voice over other types of interactions, and as voice biometrics reduces handling time by as much as 30 seconds per call and offers solid dual-factor authentication capabilities, any growth in the market will be tempered by the sheer cost of the technology itself, says speech technology consultant Donna Fluss, founder of DMG.

“The drive for biometric authentication is very high, but it’s very hard to cost-justify the technology,” Fluss says. “You have to have extremely large scale in order to do so. As it exists now, it’s a technology mainly for the companies with millions of customers.”

Like other technologies, the cost of voice biometrics will trend down, but it will do so less quickly than other technologies, Fluss adds. So movement into smaller enterprises will still be limited.

The most cost-effective way for organizations to add voice biometrics is as part of an intelligent virtual assistant (IVA) solution, according to Fluss. While she doesn’t expect the number of voice biometrics providers to change in 2019, there is coming consolidation in the IVA market.

Nuance will concentrate more of its corporate efforts on its voice biometrics business in 2019 as it completes the sale of its document imaging division to Kofax, a deal announced in November that is expected to close in the second quarter of this year. In announcing the transaction, Nuance said it would focus on its conversational AI- and cloud-based solutions while simplifying the organization and improving its growth profile.

Machine learning will figure strongly in voice biometrics, Fluss says, as companies add to their voiceprints. “Machine learning will help improve the accuracy of voice biometrics,” she adds.

That improved accuracy is critical because the attempts to compromise the technology will rise as well. Rita Singh, a research faculty member in the Language Technologies Institute at Carnegie Mellon School of Computer Science, cautions that voice biometrics is not immune from attack, saying that the technology can be thwarted in 1 out of every 100,000 attacks. With fraudsters using a variety of technologies as well as their own advancing computing technologies, the threat of attack will grow in the coming year.

So Singh expects providers to continue to enhance their own offerings by refining their voice biometric information as well as adding other parameters, such as an increasing amount of behavioral patterns. Improvement in microphones will also help in voice authentication, Singh adds. 

Phillip Britt is a freelance writer based in the Chicago area. He can be reached at spenterprises@wowway.com.

Learn more about the companies mentioned in this article in the Speech Technology Buyer's Guide:
Learn more about the companies mentioned in this article in the Vertical Markets Guide: