2019/02/18 by Jon McCormack, McCormack, Jon, Toby Gifford +9 · 3 citations
Computer Science · Neuroscience · Psychology · #FOS: Computer and information sciences #Human-Computer Interaction (cs.HC) #Music Technology and Sound Studies #Music Therapy and Health #Neuroscience and Music Perception
paper · pdf · doi:10.48550/arxiv.1902.06442
openalex publication_date 2019/02/18 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
Collaboration is built on trust, and establishing trust with a creative\nArtificial Intelligence is difficult when the decision process or internal\nstate driving its behaviour isn't exposed. When human musicians improvise\ntogether, a number of extra-musical cues are used to augment musical\ncommunication and expose mental or emotional states which affect musical\ndecisions and the effectiveness of the collaboration. We developed a\ncollaborative improvising AI drummer that communicates its confidence through\nan emoticon-based visualisation. The AI was trained on musical performance\ndata, as well as real-time skin conductance, of musicians improvising with\nprofessional drummers, exposing both musical and extra-musical cues to inform\nits generative process. Uni- and bi-directional extra-musical communication\nwith real and false values were tested by experienced improvising musicians.\nEach condition was evaluated using the FSS-2 questionnaire, as a proxy for\nmusical engagement. The results show a positive correlation between\nextra-musical communication of machine internal state and human musical\nengagement.\n