Subjective Evaluation for the Academic Text-to-Music (ATTM) Challenge

[Hao-Wen Dong]

Dear AIMC Community,

We are currently running a listening test for the ICME 2026 Grand Challenge on Academic Text-to-Music (ATTM) Generation. As the inaugural version of the ATTM challenge, we have received 16 submissions from the community and selected several finalists to enter the second-round subjective evaluation. To determine the final winners, we are soliciting volunteers from the community to participate in our listening test.

Link to the questionnaire: https://mos-listening-test-website.vercel.app/survey/127e21fe-dca6-4663-b6da-95e15633f617

In the questionnaire, you will be asked to listen to 35 10-sec music samples generated by different text-to-music models. You will be asked to evaluate each audio sample in terms of audio fidelity, prompt adherence, musicality, and overall quality. It takes around 10 minutes to complete.

We would greatly appreciate it if you can provide your input by next Monday, May 11. We apologize for the short notice due to the tight timeline for running this challenge at ICME 2026.

Your contribution will be invaluable in making this challenge successful and advancing research in this exciting field.

Please feel free to contact us if you have any questions or concerns.

Best regards,
Hao-Wen Dong, on behalf of the ATTM Challenge Organizers

--

Hao-Wen (Herman) Dong
Assistant Professor
Department of Performing Arts Technology

School of Music, Theatre & Dance
University of Michigan

hermandong.com