I thought it might be useful for some to share a good workflow I've established for myself to rapidly create different characters using an A.I based voice generating tool called Descript.

In this video you are hearing my voice (a character voice) but it is generated using only text. My use case is that I am creating a how-to-play series of videos for a boardgame I just released.

When you are creating new content at a fast pace, collaborating and that material is subject to change, the last thing you want to do is painstakingly do another recording session, or get your voice actor to do it.

The quality of the A.I generated voice is almost imperceptible from a human delivering the lines, check for yourself:

Interesting. I'm trying to figure out how it works. Did you do this with the free account?

---

It's only possible with the pro account. I think you can do it using a trial perhaps? 

The way it works is that you read half an hour of a Planet Earth script and over a couple of days the descript voice model prepares your voice for automation. Record more and get better results. Overall I read about 90 minutes to achieve the outcome you hear.