AMUSE (Advanced MUSic Explorer) - A Multitool Framework for Music Data Analysis

A large variety of research tools is available now for music information retrieval tasks. In this paper we present a further framework which aims to facilitate the interaction between these applications. Since the available tools are very different in target domain, range of available methods, learning efforts, installation and runtime characteristics etc., it is not easy to find software which is optimal for certain research goals. Another problematic issue is that many incompatible data formats exist, so it is not always possible to use output from one tool just as input for another one. At first we describe some of the available projects and outline our motivation starting the development of AMUSE framework for audio data analysis. Requirements and application purposes are given. The structure of our framework is introduced in detail and the information for efficient application is provided. Finally we discuss several ideas for further work.