On May 31st, Muzic published a research paper on their first ever text-to-midi application, MuseCoco.
Trained on a reported 947,659 Standard MIDI files (a file format which includes MIDI performance information) across six open source datasets, developers found that it significantly outperformed the music generation capabilities of GPT-4 (source).
You can explore a non-technical summary of the full collection of Muzic research papers to learn more about their efforts to train machine learning models on MIDI data.
The latest word from Pete Brown the Chair of the Executive Board of the MIDI Association, is that we will see MIDI 2.0 Windows drivers by Fall of this year.
Pete is the Principal Software Engineer and lead of the MIDI 2.0-focused Windows OS MIDI Services Project, so to subscribe to ongoing updates about Microsoft’s MIDI and related music efforts, point your browser to their DEVLOG and-or their MIDI GitHub Repository.
The full MIDI Association article and details can be found here 👇🏾