Music-01
https://minimaxi.com/en/news/music-01
An AI music generation model for synthesizing multi-style music with accompaniment and vocals.
Features
- Simultaneous generation of accompaniment and vocals
- Style learning from reference music uploads
- Lyrics-to-music generation
- Multiple genre support including classical, pop, rock, electronic, and others
- Maximum output duration: 60 seconds (3 minutes planned for next release)
Usage
- Upload reference music for style analysis
- Input desired lyrics
- Generate new music piece with learned style characteristics
Applications
- Film/TV soundtrack creation
- AI singer compositions
- Musical reinterpretations
- General music production
Technical Details
The model uses deep learning to analyze and reproduce musical patterns, rhythms, and vocal styles from reference tracks.
Limitations
- Current maximum output length: 60 seconds
- Reference track required for style learning
Future Development
- Planned support for 3-minute generation in next major release
Privacy policy
Data from this model is sent from Replicate to MiniMax.
Check their Privacy Policy for details:
https://intl.minimaxi.com/protocol/privacy-policy