
The imprisoned former Prime Minister of Pakistan, Imran Khan, has found a way to maintain his political presence despite being incarcerated since August for alleged illegal sale of state gifts.
Despite being behind bars, Khan's political party recently unveiled a four-minute video using AI-voice cloning technology to replicate his voice. Presented during a virtual rally in Pakistan, the video showcased a dubbed audio recording accompanied by a caption specifying, "AI voice of Imran Khan based on his notes."
Jibran Ilyas, a social media figure associated with Khan's party, posted the video on X. The process involved Khan sending a shorthand script to the Pakistan Tehreek-e-Insaf party (PTI), which was later refined by a legal team to match his rhetorical style. Subsequently, the text was transformed into audio reportedly using software from ElevenLabs, an AI company specialising in text-to-speech tools and voice generation.
Ilyas expressed to The Guardian, "This was a no-brainer for us when Imran Khan is no longer there to actually meet at a political rally. It was to get over the suppression."
Notably, supporters of Khan have showered the video with praise on social media platforms. It seems that Khan's social media team used archival footage from his previous rallies, combined with ElevenLab's synthetic audio technology, to replicate the politician's voice. The video was a fusion of historical footage and stock images.
The five-hour virtual rally attracted over 500,000 views on YouTube and garnered extensive viewership across various social media channels. However, to impede public access to the virtual event, the government blocked Facebook, Instagram, X, and YouTube on Sunday. NetBlocks, an internet tracking agency, confirmed a nationwide disruption to these social platforms across Pakistan on Sunday evening, as detailed in a post on X.