This endpoint allows users to replace a specific time segment of an audio clip using a textual prompt and optional lyrics. The inpainting operation blends new audio content into the selected range, guided by user-defined style and voice preferences.
To use Inpaint, upload a song and provide its full lyrics. Edit only the section in the lyrics you want to change, but keep the rest unchanged. Your edits must match the exact timestamp you select (e.g., if you pick 0:11–0:19, only update lyrics in that range).
ℹ️Experimental Feature: We do not guarantee continuous reliability and a bug-free experience as this feature is in beta.
URL or S3 path to the input audio.
"https://mybucket.s3.amazonaws.com/song.mp3"
A description of how the replacement should sound.
"Replace this part with an opera-style vocal."
Time in seconds to start replacing audio.
12.5
Time in seconds to stop replacing audio.
20
Uploaded input audio file.
Lyrics to be used for inpainting.
2000"This is where my story begins"
Voice style for the inpainted segment.
male, female, neutral "male"
Callback URL for async processing results.
"https://example.com/webhook"