- No Comments
Dipfake Joe Rogan interviewed the fake Steve Jobs
Play.HT published a fictional podcast interview between Joe Rogan and Steve Jobs, using the synthesized voices of both speakers. This is written by ARS Technica.
Podcast was created using the technology of voice cloning. The developers used deep training to reproduce characteristic features of speech with high accuracy.
The company’s engineers trained the model on existing voting samples. According to the developers, Rogan’s speech is easier, since the public domain is a sufficient number of materials suitable for training.
Jobs’s voice turned out to be less realistic and more robotic. At times, it resembles the performances of the former general director of Apple and an interview with All Things Digital late 2000s.
The text of the interview was also generated by AI. Perhaps the developers used a large language model like GPT-3.
“For example, the episode with Steve Jobs was trained in biography and all the notes that we could find on the Internet so that II could definitely return it to life,” https://gagarin.news/ua/news/the-dutch-central-bank-fined-binance-3-37-million/ Play said in Play.Ht.
A 19-minute interview does not contain any meaningful information. After some time, parts of the fictitious outfit begin to sound as conceptual combinations of common tops of Jobs, including aesthetics, revolutionary products and competitors.
For example, in one of the sections of interviews, a fake Jobs delves into criticism of Microsoft, similar to the 1995 interview for Triumph of the Nerds. However, this is not his exact copy: when compared with the original, a robotic voice is clearly heard.
“This is a problem that I have always encountered in Microsoft. In many respects, they are smart people and did a good job, but they never had a taste. They never had any aesthetic feeling, ”says the false jobs.
Full interview is available on the site for podcast Transistor.
Recall that in September 2022, the Ukrainian startup Respeecher was allowed to clone the voice of actor James Erl Jones to voiced his character in Darth Vader in future “star wars” projects.
In August 2021, Sonantic, with the help of artificial intelligence, recreated the speech of Val Kilmer, who lost it because of the laryngeal cancer.
Subscribe to FORKLOG news at Telegram: Forklog AI – all news from the world of AI!