Regenerate allows you to seamlessly edit your recorded audio by replacing words, phrases, or entire sections using AI-generated speech. This feature helps you refine your recordings without needing to re-record, making the editing process faster and more efficient.
If you're trying to modify words in a recording that isn't in English, Regenerate only works with English recordings. If you need to replace words and generate audio in another language, Translate can convert the entire script and generate a new voiceover in the selected language.
After you import and transcribe a recording or record directly in the app, AI speakers make correcting your recordings as easy as typing.
- If not already present in your script, add a speaker by typing
@
in your script. - Make sure to select an AI speaker from the list. If you’re creating a new one, make sure to enable speech generation before proceeding.
- Highlight the section of script you want to Regenerate.
- Select Regenerate in the selection bar (Or just press
D
). - Type the audio you want to generate and press Regenerate.
Give Descript a moment to generate your audio. Once your regenerated text has a green checkmark next to it, you can listen and playback the section.
When using Regenerate on a video with audio, Descript adjusts the video's speed to synchronize with the new audio. Descript will try to make the video's timing align with the updated audio.
- Stock voices are not available for use with Regenerate.
- Using Regenerate audio on existing audio clips is only available in English.
- You can replace up to 250 characters at once.
- Regenerating sections containing a video sequence will remove any video content in the selected range.