Introducing Voice Cloning (Private Beta) - Narration in your own voice
TL;DR - Voice Cloning is now in Private Beta. Type a script and have it narrated in your own voice. Your voice data is private to your account, and access is invite-only while we roll it out.
Your voice, generated from text
Record a short voice sample once. Tight Studio trains a voice model from it and saves it to your account. From then on, any project can generate narration in your voice from a script - paste the text, pick a take, drop it on the timeline.
The output carries your cadence and your tone. Patch a single sentence, replace a paragraph, or write the whole narration from scratch without ever opening a microphone.
Your voice is your data
A voice model is sensitive. We treat it that way.
- Your voice sample and trained model are private to your account. They are not shared with other users, not used to train shared models, and not exposed in any public library.
- Generated narration belongs to the project you generated it in. Nothing leaves your account unless you choose to share or export it.
- You can delete a voice model at any time, and the underlying training data is removed with it.
Voice is identity. The defaults reflect that.
Private Beta - invite-only
Voice Cloning is launching as a Private Beta. Access is invite-only while we ramp the feature up - both to keep generation quality high under load, and to make sure the controls around voice data hold up before we open it to everyone.
If you want early access, reply to the launch announcement or email [email protected] with a short note about how you’d use it. We’re sending invites in batches over the next few weeks.
What’s next
Voice Cloning closes out Launch Week 1. Five days, five shipments. Thanks for following along - and if you missed any of them, the Launch Week 1 page has the full set.