| "Voice Clone" Technology Sparks Debate over Audio Misappropriation Recent reports of artificial intelligence being used to illegally create audio of British writer and comedian Stephen Fry have sparked outrage and debate over voice clone technology. The incident has revealed both the potential of AI to manipulate sound recordings and the vulnerability of individuals whose words can be exploited by malicious actors. The misuse of audio began when a short clip was shared online using Fry’s voice saying the words, “He was a wise man who invented beer”. Fry immediately took to social media to deny the clip was actually him, and to condemn the theft of his voice. “It’s of course not me speaking,” he posted to his 3.5 million followers on Twitter last week. The reality behind Fry’s post was soon revealed. The audio was in fact created using a new technology called voice cloning. By piecing together snippets of Fry’s recorded voice, AI algorithms were able to generate a realistic-sounding copy. Voice cloning is made up of two different technologies. Firstly, a text-to-speech technology that takes text input and converts it into an audio output. And secondly, a voice cloning algorithm which uses recordings of a person to replicate their speech pattern. Voice cloning technology has been around for a while. However, this incident has highlighted the dangers of it being used for malicious purposes. It’s prompted conversations about the need to protect individuals from words and phrases being misrepresented, and who should be held accountable when audio is used unlawfully. The worrying reality is that this technology is still evolving. As it continues to be developed further, it’s essential for safeguards to be put in place to ensure it’s not used for ill gain. Unless legislators and industry leaders tackle the issue head on, this technology is likely to become an even bigger concern in the future. |