HIGH-M in detail

What is HIGH-M about?

HIGH-M (Human Interaction assessment and Generative segmentation in Health and Music)

 
Musical improvisation has proven to be an effective method of music therapy. Furthermore, it may provide a relevant approach to clinical assessment and diagnosis. Through improvisation, one can observe specifics in the beaviour of people with dementia or depression in a live manner instead of relying on self-statements in questionnaires. The HIGH-M project develops a procedure for the automated analysis of musical interaction in clinical improvisations and contributes at the time to the applicability of music therapeutic methods in clinical diagnostics and assessment.

HIGH-M (Human Interaction assessment and Generative segmentation in Health and Music) is a cooperation between the Technical University of Applied Sciences Würzburg-Schweinfurt (THWS), the AI-Juniorprofessorship of the University of Music Nuremberg and further international partners such as the RITMO Centre for Interdisciplinary Studies in Rhythm, Time and Motion (University of Oslo), the University of Jyväskylä, and the Cambridge Institute for Music Therapy Research.

In the joined scientific work, several theories for analysis are synthesised, that connect the microanalysis of musical interaction (Improvisation Assessment Profiles - Autonomy Microanalysis), data-based structures and computational visualisation (Music Therapy Toolbox - MTTB). Central questions in this procedure revolve around the identification and specificity of different types of interaction, which qualify the musical and social relation between client and therapist. To formalise those interaction types, we further integrate a sociologist game theory (Social Systems Game Theory) as a framework that also focuses on the impact of the particular situation of clinical improvisation on the musical result. 

The aim of our project is the development of an AI-assisted Tool that automatically identifies and classifies the different types of musical and social interaction, and therefore assists the effective assessment in clinical music therapy. 

Contact Persons at IFAS: Prof. Thomas Wosch (Project Lead), M.A. Bastian Vobig (PhD-Candidate)

Cooperation Partner: national cooperation with Prof. Sebastian Trump (AI-Juniorprofessorship "Artificial Creativity and musical Interaction") and Prof. Martin Ullrich (both University of Music Nuremberg), international cooperations with Dr. Olivier Lartillot and M.A. Anna-Maria Christodolou (RITMO at University of Oslo), Prof. Jaakko Erkkilä (University of Jyväskylä) as well as Prof. Jörg Fachner and Dr. Clemens Maidhoof (Cambridge Institute for Music Therapy Research)

Funding: THWS-HTA (High Tech Agenda Bayern) Förderlinie A

Further information can be requested at bastian.vobig[at]thws.de