OpenAI's Voice Tool can clone a voice from 15 seconds of recorded input

[gary]

Deemed too risky for general release by Open AI, their "new" Voice Tool can generate a convincing clone of anyone’s voice using just 15 seconds of recorded audio.

In a 29 March 2024 blog post, Open AI present some audio examples :-

https://openai.com/blog/navigating-the-challenges-and-opportunities-of-synthetic-voices

[JA]

Thx Gary.

BTW It's a great reason not to speak first when you pickup a call from an unknown caller until you can hopefully verify you are speaking to a human for a bona fide reason. Or not even pickup at all.

Best
JA

[gary]

Absolutely! Or you get a call from a "family member" asking for help.

Then just when you think all of this can only be bad, listen to the example
audio with the patient with the brain tumor.