CSIRO’s Data61 Developed Voice Liveness Detection ‘Void’ to Safeguard Users Against Voice Spoofing Attacks

Spoofing attacks that impersonate user’s devices to steal data, spread malware, or bypass access controls are becoming increasingly popular as the threat actors expand their horizon with the improvisation of various types of spoofing attacks. Especially, voice spoofing attacks that have been on a rise as more and more voice technologies are being equipped to send messages, navigate through smart home devices, shop online, or to make use of net banking.

In a joint effort for the aforementioned concern, Samsung Research and South Korea’s Sungkyunwan University and Commonwealth Scientific and Industrial Research Organisation’s (CSIRO) Data61, came up with ‘the voice liveness detection’ (Void) to keep users safe against voice spoofing attacks.

In order to detect the liveness of a voice, Void gains insights from a visual representation of the spectrum of frequencies known as ‘spectrograms’ – it makes the functionality of void a little less complex compared to other voice spoofing methods that rely on deep learning models, as per Data61.

How Void helps in detecting hackers spoofing a system? 

The void can be inserted in consumers’ voice assistance software or smartphones in order to spot the difference between ‘a voice replayed using a speaker’ and ‘a live human voice’, by doing so it can easily identify when a cybercriminal attempts to spoof a user’s system.

While giving further related insights, Muhammad Ejaz Ahmed, a cybersecurity research scientist at Data61, told, “Although voice spoofing is known as one of the easiest attacks to perform as it simply involves a recording of the victim’s voice, it is incredibly difficult to detect because the recorded voice has similar characteristics to the victim’s live voice,” he said.

“Void is a game-changing technology that allows for more efficient and accurate detection helping to prevent people’s voice commands from being misused.”