Microsoft’s new text-to-speech model can duplicate anyone's voice in 3 seconds

Re:Microsoft’s new text-to-speech model can duplicate anyone... by BatteryMountain on Monday January 09, 2023 @07:20AM (#34309521)

And Open AI's Whisper, which for me is the cherry on top!

[ Parent | on HN ]

Re:Microsoft’s new text-to-speech model can duplicate anyone... by nprateem on Monday January 09, 2023 @07:53AM (#34309774)

Jesus it can even understand Scottish. I need this app!

[ Parent | on HN ]

Re:Microsoft’s new text-to-speech model can duplicate anyone... by pjc50 on Monday January 09, 2023 @08:05AM (#34309869)

Obligatory Burnistoun reference: https://www.youtube.com/watch?v=TqAu-DDlINs

But I'd be interested to see what its reliability is like on various types of Scottish accent.

[ Parent | on HN ]

Re:Microsoft’s new text-to-speech model can duplicate anyone... by pjc50 on Monday January 09, 2023 @08:04AM (#34309860)

I tried Whisper on an admittedly very challenging task (live translate Albanian to English subtitles), and it failed miserably. Went back and saw it had a 44% error rate for that language.

Might give it another go on an easier task.

[ Parent | on HN ]

Re:Microsoft’s new text-to-speech model can duplicate anyone... by lajamerr on Monday January 09, 2023 @12:55PM (#34313409)

Even though Whisper advertises the translation part of it. I honestly wouldn't say it's a first class feature at least part of the Whisper model itself.

The only thing I'd use Whisper for is transcription part of it. Then use ChatGPT for the translation.

[ Parent | on HN ]

Re:Microsoft’s new text-to-speech model can duplicate anyone... by BatteryMountain on Tuesday January 10, 2023 @01:57AM (#34321435)

Yeah the top 5 languages on that WER graph is very much worth a look, the rest a bit error prone.

Surprisingly, I tested it with my mother, she has a very broken english accent/dialect and it worked fine for her. Works amazing actually and I'm busy building some tools around it for her to test.

[ Parent | on HN ]