The Coast Guard is using voice recognition to combat prank calls

April 28, 2017 - 

The National Coast Guard is implementing voice recognition technology to combat the rise in fake distress calls that have been flooding the agency’s VHF radio channel in recent months, according to a report by The Verge.

The agency is obligated to respond to every call considering there is a potential life at risk, which can be costly and time-consuming when it’s all for a childish joke.

“We’re just getting more hoaxes every day,” says Lt. Gianfranco Palomba, who’s been tasked with stopping the calls. “We’re seeing a direct impact, not just on man-hours but on assets.”

The Coast Guard believes there are a few pranksters making up to 10 calls each through the VHF radio channel, which is the maritime emergency communications line, picked up by a network of antennas along the coast.

Unfortunately, there is no phone number or any identifying information to trace back a caller, and the Coast Guard’s network of VHF antennas aren’t nearly dense enough to triangulate the source of a call.

Palomba and his team is testing voice recognition technology to identify the pranksters, which gives a clear sample of the subject’s voice regardless of how little he or she says on the call.

The software could enable the Coast Guard to match the same caller the next time they phone in a hoax.

Nuance offers software that can build a voiceprint from only 40 seconds of speech, however, many of the fake distress signals tend to be much shorter so it remains to be seen if the software can help the Coast Guard.

“We all speak differently, not just because of the size and shape of our larynx, but because of where we grew up, because of our personality,” Nuance engineer Brett Beranek said. “That makes voice biometrics a little more complex, but that complexity gives us a lot more data to play with.”

Carnegie Mellon University professor Rita Singh, who has been examining the Coast Guard’s hoax calls since 2014 under a separate research partnership, said he can reliably tell the caller’s gender and age. However, extracting any additional information from a call can be challenging..

Many of the callers seem to use a different voice to conceal their identity, which only makes the identification process more difficult for any automated voice-readers.

“Although probably not aware of the biometric potential of their voice, they instinctively attempt to hide their identity by disguising it,” Singh wrote in a recent paper. “They try to sound like a real (albeit fictitious) person other than themselves.”

The Coast Guard has yet to find a commercial system that can identify any voice previously used to make a hoax call, and is still seeking recommendations and best practices from academia and other government agencies.

In the meantime, Palomba is looking to undergo a high-profile arrest and trial process to discourage pranksters from making fake distress calls.

“People still see that wiggle room, just because they haven’t been reminded of a really big prosecution lately,” Palomba said. “We’re looking for that deterrent effect.”

Leave a Comment

comments

About Justin Lee

Justin Lee has been a contributor with Biometric Update since 2014. Previously, he was a staff writer for web hosting magazine and website, theWHIR. For more than a decade, Justin has written for various publications on issues relating to technology, arts and culture, and entertainment. Follow him on Twitter @BiometricJustin.