Microsoft made an AI voice so real, it’s too dangerous to release

Microsoft has a brand new AI toy — and no, you’ll be able to’t use it.

Researchers on the firm revealed in a analysis paper (noticed by Dwell Science) that they’ve constructed an AI voice synthesizer that may create plausible, sensible human voices straight from textual content prompts.

Referred to as VALL-E 2, the analysis paper claims it is the primary time “human parity” has been achieved in a text-to-speech system.

After all, there is not any means for any of us to know that for certain as a result of common individuals cannot use it. Microsoft’s weblog submit may be very clear about this.

“This web page is for analysis demonstration functions solely,” Microsoft’s web site stated. “At present, we’ve got no plans to include VALL-E 2 right into a product or increase entry to the general public.”

Sadly, there aren’t audio samples out there, so you may should belief the researchers on this. Nonetheless, the weblog submit does supply quite a few charts and technical phrases to discover. If it is nearly as good as they are saying, although, possibly it is for the most effective that we won’t hear it.