Microsoft's Azure AI Speech needs just seconds of audio to spit out a convincing deepfake

Microsoft has upgraded Azure AI Speech so that users can rapidly generate a voice replica with just a few seconds of sampled speech. The personal voice feature for AI Speech became generally available on May 21, 2024. It was impressive but required some training to get the best out of it. According to Microsoft, the feature …

[Anonymous Coward]

All customers must agree to our usage policies

Is that the bit where you click on the fire hydrants? Did that months ago.

[Dan 55]

Wonderful

Can't wait to see what product Big Tech has up its sleeve next with no purpose other than to ruin society.

[druck]

Re: Wonderful

Well you can't expect Microsoft not to want to take a cut of all those impersonation and SOS scams.

[O'Reg Inalsin]

And Wells Fargo is still promoting voice recognition as a safe IDSecurity option.

[druck]

And HSBC; every time you use telephone banking they encourage you to "make your voice your password".