Skip to content

JASCO

Text-to-music with local and global controls

Visit official site

What JASCO is

The model is described as a text-to-music system that joins symbolic and audio-based conditions, shaped by a global text description.

What you can do with it

  • Generate high-quality music samples from text
  • Apply local controls such as chords
  • Combine symbolic and audio conditions

Pros & cons

✓ What we like

  • Chords can act as a local control
  • The material states more flexible control than baseline models

! What to watch out for

  • Control is split between global and local inputs
  • Quality is evaluated by objective metrics and human studies

FAQ

What conditions are combined?

Symbolic and audio-based conditions in one model.

What local controls are named?

The documented steps name chords or melody.

Last reviewed: 2026-11-15

More AI music creation tools

View all →

How we review