Create a deepfake of your own voice with this podcast tool

Deepfakes are a major security concern for their ability to make it appear as though people are doing or saying something they never actually did, but one startup is trying to use the artificial intelligence technology for something simpler: Podcast editing. Descript has launched a new feature in its podcast software called Overdub in beta on Wednesday, allowing users to create a realistic text-to-speech model of themselves by uploading a few minutes of audio.

The technology, which comes from partner company Lyrebird, is meant to save podcasters from having to re-record or splice audio when a mistake or change is made. Instead, you can simply type the words you want to add into the recording, and Overdub will make it sound as if your voice is saying them.

It is expressly not meant to create deepfakes, according to a blog post from Andrew Mason, Descript CEO. To train the voice model, users need to record themselves speaking randomly generated sentences -- so others can't use pre-existing recordings to create a model using someone else's voice, the post said.

However, Descript acknowledged in an ethical FAQ that while its product is unique, the foundational research is already widely available. Future products may not have the same constraints for use.

"That's why it's important for us to showcase the technology to the world in a controlled environment," Jose Solteo of Lyrebird said in an email to CNET. "So that the world can be better prepare against potential malicious attacks."

The company stores the recordings used to create a voice double. However, users can delete their recordings at any time, Solteo said.

Deepfakes have targeted everyone from Mark Zuckerberg to Tom Cruise and Vladmir Putin, and even the fictional Ron Swanson and the cast of Full House. US lawmakers have said deepfakes "could pose a threat to United States public discourse and national security."