Panos Papiotis

The Sense of Ensemble: a Machine Learning Approach to Expressive Performance Modelling in String Quartets

Computational approaches for modelling expressive music performance have produced systems that emulate music expression, but few steps have been taken in the domain of ensemble performance. In this paper, we propose a novel method for building computational models of ensemble expressive performance and show how this method can be applied for deriving new insights about collaboration among musicians. In order to address the problem of inter-dependence among musicians we propose the introduction of inter-voice contextual attributes. We evaluate the method on data extracted from multi-modal recordings of string quartet performances in two different conditions: solo and ensemble. We used machine-learning algorithms to produce computational models for predicting intensity, timing deviations, vibrato extent, and bowing speed of each note. As a result, the introduced inter-voice contextual attributes generally improved the prediction of the expressive parameters. Furthermore, results on attribute selection show that the models trained on ensemble recordings took more advantage of inter-voice contextual attributes than those trained on solo recordings.

Measuring ensemble interdependence in a string quartet through analysis of multidimensional performance data

In a musical ensemble such as a string quartet, the musicians interact and influence each others actions in several aspects of the performance simultaneously in order to achieve a common aesthetic goal. In this article, we present and evaluate a computational approach for measuring the degree to which these interactions exist in a given performance. We recorded a number of string quartet exercises under two experimental conditions (solo and ensemble), acquiring both audio and bowing motion data. Numerical features in the form of time series were extracted from the data as performance descriptors representative of four distinct dimensions of the performance: Intonation, Dynamics, Timbre, and Tempo. Four different interdependence estimation methods (two linear and two nonlinear) were applied to the extracted features in order to assess the overall level of interdependence between the four musicians. The obtained results suggest that it is possible to correctly discriminate between the two experimental conditions by quantifying interdependence between the musicians in each of the studied performance dimensions; the nonlinear methods appear to perform best for most of the numerical features tested. Moreover, by using the solo recordings as a reference to which the ensemble recordings are contrasted, it is feasible to compare the amount of interdependence that is established between the musicians in a given performance dimension across all exercises, and relate the results to the underlying goal of the exercise. We discuss our findings in the context of ensemble performance research, the current limitations of our approach, and the ways in which it can be expanded and consolidated.

The Plausibility of a String Quartet Performance in Virtual Reality

We describe an experiment that explores the contribution of auditory and other features to the illusion of plausibility in a virtual environment that depicts the performance of a string quartet. ‘Plausibility’ refers to the component of presence that is the illusion that the perceived events in the virtual environment are really happening. The features studied were: Gaze (the musicians ignored the participant, the musicians sometimes looked towards and followed the participants movements), Sound Spatialization (Mono, Stereo, Spatial), Auralization (no sound reflections, reflections corresponding to a room larger than the one perceived, reflections that exactly matched the virtual room), and Environment (no sound from outside of the room, birdsong and wind corresponding to the outside scene). We adopted the methodology based on color matching theory, where 20 participants were first able to assess their feeling of plausibility in the environment with each of the four features at their highest setting. Then five times participants started from a low setting on all features and were able to make transitions from one system configuration to another until they matched their original feeling of plausibility. From these transitions a Markov transition matrix was constructed, and also probabilities of a match conditional on feature configuration. The results show that Environment and Gaze were individually the most important factors influencing the level of plausibility. The highest probability transitions were to improve Environment and Gaze, and then Auralization and Spatialization. We present this work as both a contribution to the methodology of assessing presence without questionnaires, and showing how various aspects of a musical performance can influence plausibility.

Prototyping interactions with Online Multimodal Repositories and Interactive Machine Learning

Interaction designers often use machine learning tools to generate intuitive mappings between complex inputs and outputs. These tools are usually trained live, which is not always feasible or practical. We combine RepoVizz, an online repository and visualizer for multimodal data, with a suite of Interactive Machine Learning tools, to demonstrate a technical solution for prototyping multimodal interactions that decouples the data acquisition step from the model training step. This way, different input data set-ups can be easily replicated, shared and experimented upon their capability to control complex output without the need to repeat the technical set-up.