Prediction of real-time musical interactions in networked music performance
keywords MACHINE LEARNING
Research Groups Telecommunication Networks Group
Thesis type SIMULATION/ANALYSIS
Description Prediction of real-time musical interactions in networked music performance
Networked Music Performance aims at revolutionizing the traditional concept of musical interaction by enabling remote musicians to perform together through a telecommunication network. Ensuring realistic performative conditions, however, constitutes a significant engineering challenge due to the extremely strict requirements of audio quality and network latency, which are rarely met by current internet technologies. Audio data packets are, in fact, often lost or received too late, leading to unacceptable degradation of the musical quality.
The objective of this thesis project is a preliminary feasibility analysis of the application of Machine Learning (ML) methods to develop predictive algorithms capable of foreseeing the performance of a musician ahead enough to compensate for communication delays and of anticipating expressive changes in order to preserve the impression of an interactive performance. The project activities include the implementation of a proof of concept prediction algorithm leveraging state-of-the art ML methods such as artificial neural networks and to evaluate its performance in terms of prediction error and perceived quality of experience.
Required skills Programming skills
Deadline 08/01/2020 PROPONI LA TUA CANDIDATURA