Author: Oded Ben-Tal

Interpreting Data through Creative Audiovisualisation (IDCA) – Progress, pitfalls and potential ways forward.

This talk presents work currently taking place as part of an RSE-funded Fellowship exploring the potential of simultaneous audiovisualisation as a means of both data exploration and compositional structuring. It discusses three of the specific datasets currently being explored, the outcomes of those explorations and the challenges of balancing the requirement for clarity in data presentation with the want and need to create engaging pieces of audiovisual work. Ultimately, it poses the question “is it possible to create audiovisual works that both communicate data and are aesthetically satisfying for the composer?”.

For the first seminar of 2023 we are delighted to welcome Professor Devaney (Brooklyn College). She will be speaking about Integrating expert domain knowledge into computational models of musical understanding.

Humans are able to learn with greater efficiency than machine learning models, in large part because they learn not just from exposure, but also from domain knowledge, which includes codified knowledge and guided practice. Examples in the musical domain are established pedagogies found in music theory textbooks and the skilled interpretative practice found in musical performances. In music information retrieval (MIR), the music understanding process has been approximated by individual tasks such as chord recognition, rather than treating such single-label classification tasks as only one step in analysing musical signals. This talk discusses ongoing work of how, in the Western Art music domain, chord recognition can be used to parse the musical surface into structural and non-structural components using not just exemplar-based learning but also domain knowledge derived from music theory textbooks and music performance data. 

On Thursday, December 1st, we welcome professor Ian Cross (Cambridge University) to our virtual seminar. He will be talking about Music, Speech and Affiliative Communicative Interaction: Pitch and Rhythm as Interactive Affordances

Abstract: This paper presents the idea that, across cultures, musical interaction is communicative and intrinsically affiliative, overlapping significantly in function with the phatic speech register.  I will refer to experimental evidence that supports the view that joint music-making and phatic conversation can overlap not just in function but also in form.  I will suggest that what can be interpreted as phatic conversation or as participatory music are manifestations of a superordinate domain of affiliative communicative interaction, and will submit that features that we tend to think of as musical —such as discrete patterns in pitch and rhythm— are best construed not as aesthetic properties but as interactive affordances that may be functional in the achievement of interpersonal affiliative alignment.

In two weeks we will be in Stockholm hosted by Dr. Bob L.T. Sturm at KTH/KMH. The workshop will include two concerts, keynotes from Prof Marc Leman and Cecilia Åsberg, short talks and discussions. The workshop is co-located with the MUSAiC festival. If you are in Stockholm and wish to join us please email info@musicdatasense.org.

Prediction learning is considered a ubiquitous feature of biological systems that underlies perception, action, and reward. For cultural artifacts such as music, isolating the genesis of reward from prediction is challenging, since predictions are acquired implicitly throughout life. Here, we examined the trajectory of listeners’ preferences for melodies in a novel musical system, where predictions were systematically manipulated. Across seven studies (n = 842 total) in two cultures, preferences scaled with predictions: participants preferred melodies that were presented more during exposure (global predictions) and that followed schematic expectations (local predictions). Learning trajectories depended on music reward sensitivity. Furthermore, fMRI showed that while auditory cortical activity reflects predictions, functional connectivity between auditory and reward areas encodes preference. The results are the first to highlight the hierarchical, relatively culturally-independent process by which predictions map onto reward. Collectively, our findings propose a novel mechanism by which the human brain links predictions with reward value.

Monday 27/6 at 4pm (UK)