The team Data61 of Commonwealth Scientific and Industrial Research Organization (CSIRO)In Australia, along with Samsung Research and South Korean Sungkyunkwan University, have designed and developed a solution for it protection of users from vocal spoofing attacks.
This solution is called “Voice liveness detection” (Void) and as its name suggests it is designed to detects if the sound being heard actually comes from a human. It is designed to incorporated into smartphone or in voice assistance software and is able to detects the difference between a "live" human voice and a speaker, a tactic often followed by hackers to represent someone else and fool them systems.
According to Data61, Void uses deep learning models to detect vocal spoofing and is based on information from spectroscopes, a visual representation of the frequency spectrum to detect if the voice is indeed "alive".
The CSIRO Data61 researcher, Muhammad Ejaz Ahmed, explained that spoofing attacks become more and more common as vocal technologies are now used for purchases in Internet, making phone calls, sending messages, checking smart home devices and access in banking services.
"Although voice spoofing is known as one of the easiest attacks, as it simply involves recording the victim's voice, it is incredibly difficult to locate because the Recorded voice has similar characteristics to real voice of the victim ", he said.
"Void is a game-changing technology that allows for more efficient and accurate detection, which helps prevent people from abusing people's voice commands."
In the context of Void development, many tests have been performed. According to Data61, the new technique has been able to pinpoint almost all of them scams. In addition, he identified attacks, eight times faster than other deep learning techniques.
The results of the tests have been published in a research paper entitled, "Void: A fast and light voice liveness detection system", which will be presented at a conference in August.