Create an AI voice clone of a guest

Create a reusable AI version of a guest's voice from a short video they record, then generate audio from any script you write.

Plan: Business
Device: Computer browser

This feature uses AI Credits

Step by step

Navigate to the relevant studio.
In the sidebar, click Voice clones.
If this is your first voice clone, the setup options appear on the page.
If you're creating an additional voice clone, click + New voice clone.
Select Someone else's voice.
Under Their name, enter your guest's name.
Your guest's name fills in the consent statement automatically.
Review the script under What they'll read on camera. Your guest reads this script on camera, including a statement giving you and Riverside permission to create their AI voice clone.
To invite your guest:
- By email: enter their email address under Their email, then click Send invite.
- By link: next to Not sure of their email?, click Copy invite link, then share the link with your guest.
Click Done.
Your guest's voice clone appears on the Voice Clones page with the status Waiting for consent until they record.
Your guest opens the invite, enters their details, checks their camera and mic, and reads the script from the teleprompter.
They can record again as many times as they need before submitting.
When the voice clone is ready, click Generate track on its card.
Type or paste your script (up to 5,000 characters).
Click Generate track.
You'll get an email when your video is ready.

Your guest doesn't need a Riverside account. They record in their browser from the invite link.
Riverside trains your guest's voice clone only on the audio from their invite session.
Riverside saves your guest's consent recording with their voice clone.
You can't generate audio with your guest's voice clone until they record their consent video.
Your guest can't change the script on the teleprompter.