Create custom voice clones from audio samples for use in your AI-generated music.
Music AI includes a Voice Library with 50+ pre-built AI voices. Additionally, you can create custom voice clones by uploading audio samples through the Singer Studio.
The current voice synthesis system uses Bark TTS, which is excellent for rap and speech-style vocals. True singing synthesis is coming in a future update!
High-quality voice clones start with high-quality samples. Here's what you need:
| Specification | Requirement |
|---|---|
| Duration | 10-30 seconds of clear audio |
| Content | Speech or acapella vocals (no background music) |
| Quality | Clear recording without background noise |
| Format | MP3, WAV, or other common audio formats |
Use acapella tracks or spoken word recordings without any background music or effects. The cleaner the sample, the better the clone.
For best results, your sample should include:
Music AI includes 50+ pre-built AI voices optimized for different genres:
Female | EN, DE, ES | Fast flow, aggressive female rap
Male | EN, DE | Aggressive gangsta rap style
Male | DE | German street rap specialist
Male | EN | Aggressive delivery, battle rap
Female | EN, ES | Pop ballads, catchy hooks
Female | EN | Pop, Broadway style
Pair your custom voice with a complementary pre-built voice. For example, use your voice for verses and a polished Pop Star voice for choruses.
Voice cloning requires explicit consent from the person whose voice is being cloned. Unauthorized voice cloning may violate laws and our Terms of Service.
Cloning your own voice for personal or commercial projects is always permitted and encouraged!
Create a 10-30 second clear audio recording of the voice.
Go to Singer Studio and upload your sample to create the clone.
Enable vocals and choose your custom voice from the dropdown.
Generate tracks, listen, and adjust settings as needed.