Fourth Annual Machine Folk Music School

Organisers: Bob L. T. Sturm (KTH Royal Institute of Technology)

Bio: Sturm has been experimenting with machine folk music since building a folk-music-focused machine learning model in 2015 “for a laugh”. He runs the blog, Tunes from the Ai Frontiers, which features performances of new machine folk tunes and other artifacts of human-Ai collaboration. In 2020, Sturm founded The Society for the Preservation and Promotion of Machine Folk Music (v1.1), with support of the MUSAiC project (Music at the Frontiers of Artificial Creativity and Criticism, ERC-2019-COG, No. 864189). One of the remits of The Society is the organization of Machine Folk Music Schools – which have appeared at AIMC 2020, 2021 and 2022. Sturm has also been learning real Irish traditional music and accordion since 2018.

Description: This two-hour tutorial introduces participants to Ai-generated (machine) folk music through practice in person. In the first 45 minutes, one Ai-generated folk tune will be taught by the organiser, and discussed with the attendees. The tune will first be performed. Then it will be taught gradually by repeating small phrases and combining them to form the parts. Participants should be comfortable with their musical instrument of choice and be able to learn by ear (but music notation will be provided). Following a 15 minute break, the next 45 minutes presents the state of the art in the modeling of folk music using machine learning. This includes a look at sequence modeling systems applying autoregression via recurrent neural networks (LSTM) and attention (transformers), as well as masked language modeling.

Technical requirements: computer projection and sound

Setup details: An accessible classroom to accommodate 20 people with chairs.