Changing a Voice-Over into a Target Voice
This workflow is a speech-to-speech conversion.
Changing a Voice-Over into a Target Voice
This workflow is a speech-to-speech conversion.
You start with an existing audio clip, choose one of your saved voices as the target voice, and generate a converted version of that clip.
What you need first
Before you can do this, you need:
- a source audio clip already on the Video timeline
- at least one saved voice from the Voice Designer
If you do not already have a saved voice, create one first in the Voices area.
Where this happens
Voice conversion happens in the Video tab. Open the audio clip menu and choose the modify flow for that clip. In the UI, this appears as Voice Change.
What you can control
The speech-to-speech conversion flow lets you choose:
- the target voice
- the speech-to-speech model
- whether to remove background noise before conversion
How to change the voice of an existing audio clip
- Open the Video tab
- Find the source audio clip you want to convert
- Open the clip's modify flow
- Choose the saved target voice
- Choose the speech-to-speech model
- Turn on Remove Background Noise if the source recording is messy
- Generate the converted audio
- Review the returned result on the timeline
What this workflow is best for
Use voice change when:
- the performance timing is already good
- the spoken wording is already correct
- you mainly want the voice identity to change
If the source performance is poor, noisy, or badly paced, conversion alone will not fully solve that.
What to do next
Once the converted voice clip is ready, you can align it with a video clip and use LipSync in the Video Timeline.