Seed Audio Voice Cover
Create a voice cover
Train your voice, upload a song, and render a new version sung in your voice.
1Create voice model
2Upload song
3Choose voice
4Generate cover
1
Create your voice model
Upload 30 seconds to 3 minutes of clean speech or unaccompanied singing.
2
Upload the song
Choose the MP3 or WAV you are authorized to cover. The source must be 8 minutes or shorter.
3
Choose the singing voice
Your trained voice is applied as a V5 voice persona.
My Voice
RecommendedComplete step 1 to unlock My Voice.
4
Generate your cover
Seed Audio keeps the source melody and arrangement while replacing the vocal with your voice model.
Source song
Extract melody
Replace vocal
Preserve voice
New cover
Voice model readySong readyLyrics readyPermissions confirmed
Generation usually takes a few minutes and returns two versions.