Structuring Music-Related Movements

Background

A major obstacle to music-related movement research is the lack of appropriate formats and standards. Working with hardware and software tools that cannot efficiently interchange information is a limitation not only for the single researcher, but it also effectively blocks collaboration between researchers and institutions. Rather than analysing data, this makes researchers spend a lot of time and effort just getting their data transferred from one system to another. Developing formats that can be used in music-related movement research therefore seems of great importance, and is something that can also benefit other fields, e.g. music technology and human-computer interaction.

Data streaming and storage has been a major challenge in our research group for a long time. Only storing data from a single sensor interface, and synchronise these data with simultaneously recorded audio and video files is not a trivial task. In the beginning, we stored raw sensor data in text files, with references to the corresponding audio and video files. However, since we did not store any information about the setup, scaling and filtering used, etc., these data files quickly became useless as we continued to change our experimental setups. In later studies, we were more careful about taking notes, and including these with the data sets, but we still ended up with lots of problems relating to synchronisation between the data files and the audio and video files.