CARNEGIE MELLON (US) — Computer users know to preserve their privacy by safeguarding passwords, but with the advance of voice authentication systems, protecting unique voice characteristics is going to be just as important.
New technology from researchers at Carnegie Mellon University’s Language Technologies Institute (LTI) will allow people to register or check in on a voice authentication system, without their actual voice ever leaving their smartphone, reducing the risk that voice biometric data could be stolen and used later to access bank, health care, or other personal accounts.
“When you use a speaker authentication system, you’re placing a lot of faith in the system,” says Bhiksha Raj, an associate professor of language technologies. “It’s not just that your voiceprint might be stolen from the system and used to impersonate you elsewhere.
“Your voice also carries a lot of information—your gender, your emotional state, your ethnicity. To preserve privacy, we need systems that can identify you without actually hearing your voice or even keeping an encrypted record of your voice.”
Raj and Manas Pathak, a recent PhD graduate of the LTI, have devised a method for converting a voiceprint—a spectrogram that represents the acoustic qualities of speech—into alphanumeric strings that can serve as passwords.
Because a person’s voice never sends the same signal twice, even when repeating the same word or phrase, converting the voiceprint into a single password won’t do. Instead, the new system uses different mathematical functions to generate hundreds of alphanumeric strings.
To authenticate the user, the system compares all of the strings with those that the system has on file from the initial registration; if enough of the strings match, the user is authenticated.
The system also adds what the researchers call “salt”—a random string of digits unique to each smartphone—to the alphanumeric strings to provide an additional level of security. In tests using standardized speech datasets, Raj and Pathak found that their system was accurate 95 percent of the time.
The privacy-preserving method is computationally efficient, so it could be used with most smartphones, they say.
But Raj also warns that improving the security of voice authentication systems would be just a first step to protecting privacy overall. “With increasing use of speech-based services, such as the iPhone’s Siri assistant or personal videos uploaded to YouTube, the issue of the privacy of users’ speech data is only just beginning to be considered,” he says.
In addition to Raj and Pathak, Jose Portelo, and Isabel Trancoso of INESC-ID in Lisbon, Portugal, contributed to this research. This work was supported by the National Science Foundation and Portugal’s Foundation for Science and Technology (FCT).
Source: Carnegie Mellon University