Barclays Wealth is one of a growing number of institutions that has recently deployed voice recognition technology to identify customers when they call for support. With a lot of money at stake and growing sophistication of hackers, passwords alone do not provide enough protection.
A solution that has attracted a lot of interest in the banking and investment industries is FreeSpeech, a vocal biometrics package developed by Burlington, Mass.-based Nuance Communications (News - Alert).
When a business signs up a new customer, that person’s voice is examined by FreeSpeech, which creates the vocal equivalent of a fingerprint known as a voiceprint and stores it. In subsequent calls, FreeSpeech compares the caller’s voice against the voiceprint. The voice is also compared against a voice database of known scammers.
The result of FreeSpeech’s voice analysis is given to the customer service rep. If it finds that the voice does not match the voiceprint saved in the database, an alert is issued. The rep can then ask for additional identifying information to validate the customer.
In addition to Barclays, several financial institutions in India, American financial institutions like Vanguard and Eastern Bank and Turkish mobile phone service Turkcell (News - Alert) are adopting Nuance voice biometrics.
Voice biometrics technology is popular and expected to continue to be popular for the next five years. According to Opus Research, spending on the technology increased 74 percent between 2011 and 2012 and will grow at a compound annual growth rate of 35 percent over the next five years.
In an interview with Computing, Matt Smallman, vice-president of global client services for Barclays touted the benefits of FreeSpeech:
“Our frontline people have historically had the weight of this whole decision [of the validity of a caller] on them, and have had little support to make the decisions successfully. You could hear that they weren’t listening or concentrating on the first few minutes of the call, they were just ‘getting through security’.”
According to Smallman, only five percent of callers had to be asked for passwords when Nuance could not match their voice with a stored voiceprint.
So is voice biometrics as a security measure reliable enough to prevent fraud? By itself, it would not appear to be. If Barclays’ numbers are consistent with all customers examined by FreeSpeech, the technology fails to identify five percent of callers. It’s conceivable that a scammer with passwords and other authenticating information a rep would ask for on follow-up could still get through the system.
As part of a multiple layer security protocol, the technology appears to be an effective tool. Combining different points of information such as passwords, the phone number the call comes from as well as the voiceprint would together provide more reliable authentication.
If the technology becomes more affordable and the reliability can improve out to ‘multiple nines’, voice biometrics may make passwords, remembering your dog’s name from 40 years ago and other current security techniques obsolete.
Edited by Rory J. Thompson