||Detection of Guitar Playing Techniques Using Frequency Spectrum Analysis
||Pedersen, Christian Højgaard
||Rose, Michael (Image Analysis and Computer Graphics, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
||Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
||Music video games such as Guitar Hero and Rock Band have grown immensely popular worldwide primarily in the past decade. But toward the end of 2008 and early 2009, sales plummeted as a result of oversaturation and lack of innovation. To help reinvent the genre, games are beginning to surface that use controllers that are hybrids between traditional controllers and actual music instruments. In this report, methods for the detection of guitar playing techniques are examined using Fourier spectrum analysis, for the purpose of causing a music video game based on the guitar to react to their use. The project entails the development of a software tool that captures and transforms audio input to perform frequency analysis. The application also includes a simple front-end that displays the output and offers a few control options.
Transforming an electric guitar into a digital game controller is no simple task, and the report closely explains the difficulties inherent to polyphonic detection. Although reliable spectrum characteristics could be determined for a number of techniques, the FFTs shortcomings in terms of resolution are also among the conclusions of the project. It was found that the FFT shows promising results as a pitch detector for a guitar game, but that its limitations place certain demands on the game design. This counts for the detection of specific guitar playing techniques as well.
||Thesis not publicly available
||Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Creation date: 2010-09-20
Update date: 2010-09-20