The Music-Generative Usable+ AI (MusGU+) framework is a musician-centered evaluation framework designed to assess how generative music models can be adapted, used, and controlled in real-world creative contexts. The framework evaluates models along three complementary dimensions, with each dimension addressing a key question from the musician's perspective:
📖 Read the detailed evaluation criteria
| Model ▴▾ | Adaptability ▴▾ ≥60% | Usability ▴▾ ≥60% | Controllability ▴▾ ≥60% | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Hardware Requirements | Dataset Size | Adaptation Pathways | Technical Barriers | Model Redistribution | Interface Availability | Access Restrictions | Real-time Capabilities | Workflow Integration | Output Licensing | Community Support | Conditioning Inputs | Time-Varying Control | Feature Disentanglement | Control Parameters | |
DDSP-VST Google Magenta | ~ | ✔︎ | ~ | ~ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ✔︎ |
AFTER IRCAM | ~ | ~ | ✔︎ | ✘ | ~ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ~ | ~ | ✔︎ | ✔︎ | ✔︎ | ✔︎ |
Neutone Morpho Neutone Inc. | ✔︎ | ✔︎ | ~ | ✔︎ | ✘ | ✔︎ | ~ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ~ | ~ | ~ | ✔︎ |
RAVE IRCAM | ~ | ✔︎ | ✔︎ | ~ | ~ | ✔︎ | ✔︎ | ✔︎ | ✔︎ | ~ | ✔︎ | ~ | ~ | ~ | ✔︎ |
YuE The Hong Kong University of Science and Technology (HKUST) and MAP | ✘ | ✘ | ✔︎ | ✘ | ✔︎ | ~ | ✔︎ | ✘ | ✘ | ✔︎ | ✔︎ | ~ | ~ | ~ | ~ |
JAM Singapore University of Technology and Design and Lamda Labs. | ~ | ✘ | ✔︎ | ✘ | ~ | ~ | ✔︎ | ✘ | ✘ | ~ | ~ | ~ | ~ | ~ | ~ |
Stable Audio Open Small Stablility AI | ~ | ✘ | ✔︎ | ✘ | ~ | ~ | ~ | ~ | ✘ | ~ | ✔︎ | ~ | ~ | ✘ | ~ |
MusicGen Meta AI | ✘ | ✘ | ✔︎ | ✘ | ~ | ~ | ✔︎ | ✘ | ✘ | ✔︎ | ~ | ~ | ~ | ~ | ~ |
Suno Suno, Inc. | ✘ | ✘ | ✘ | ✘ | ✘ | ✔︎ | ~ | ~ | ✘ | ~ | ✔︎ | ~ | ✘ | ✘ | ~ |
Udio Udio | ✘ | ✘ | ✘ | ✘ | ✘ | ✔︎ | ~ | ~ | ✘ | ✘ | ✔︎ | ~ | ✘ | ✘ | ✘ |
The MusGU+ framework evaluates models across 15 criteria distributed among three dimensions: Adaptability (5 criteria), Usability (6 criteria), and Controllability (4 criteria). Each criterion is evaluated on a three-level scale: ✔︎ fully supported, ~ partially supported, or ✘ not supported.
The table includes interactive elements:
For a detailed breakdown of each model's evaluation, explore the corresponding YAML file in the projects folder.
MusGU+ builds on insights from the MusGO framework. MusGO (Music-Generative Open AI) is an openness-focused evaluation framework for music-generative AI. While MusGO assesses transparency and responsible research practices, MusGU+ supports informed selection and practical adoption of generative music models by musicians.