Paper
Al-Ghawanmeh F., Scott M. J., Menacer M. and Smaili K., 2021. Predicting and Critiquing Machine Virtuosity: Mawwal Accompaniment as Case Study. The International Computer Music Conference 2021.
Summary
I believe we have moved beyond the question of whether or not the machine can compose decent music. The question today is whether the machine can really become a virtuoso! And, if yes: how, to what cost, and when?
In this contribution to ICMC21, we tried to answer these questions in the context of maqam music, the traditional music of the Middle East and North Africa. We conceptualized virtuosity within the context of tarab, or modal ecstasy, and assessed the extent to which machine virtuosity can elicit ecstasy among musicians and audiences. We chose the mawwal, a deep-rooted vocal improvisation, and presented a responsive melodic accompaniment based on statistical machine translation. So, in other words, we translated vocal music into an instrumental response.
On our journey to understand the machine's potential for virtuosity, we applied an objective evaluation over three phases of corpus expansion. We then followed it with a subjective evaluation and a prediction for the machine’s future performance. Finally, we explored a critique of the machine's virtuosity through a user-experience study.
Bio
My passion is the design and development of high quality computer-generated (or assisted) music influenced by maqam music. I also aim to spread this knowledge via education, performance, and multimedia. I apply Natural Language Processing techniques for AI-based music composition, and also do audio and music signal processing.
I have enjoyed developing Mawaweel, an application that provides automatic accompaniment customized to Arab music. I started with a knowledge-based model when completing my MA in Music Technology at New York University. I continued my research when I joined the University of Jordan’s Music Department as an instructor. In the past period, I started investigating a machine learning approach based on Statistical Machine Translation. This led me to my current PhD research at the University of Oslo and the University of Lorraine, where I am working on automatic composition based on Natural Language Processing with control to allow for individual customization.