Research

Non-thesis publications

  • Physical Modeling meets Machine Learning: Teaching Bow Control to a Virtual Violinist. Graham Percival, Nicholas Bailey, George Tzanetakis, Sound and Music Conference 2011. pdf
  • MOGFUN: musical mObile group for FUN. Yinsheng Zhou, Graham Percival, Xinxi Wang, Ye Wang, and Shengdong Zhao, CHI 2011. pdf *Honorable mention award*
  • MOGCLASS: a collaborative system of mobile devices for classroom music education. Yinsheng Zhou, Graham Percival, Xinxi Wang, Ye Wang, and Shengdong Zhao, ACM Multimedia 2010. pdf
  • MOGFUN: musical mObile group for FUN. Yinsheng Zhou, Zhonghua Li, Dillion Tan, Graham Percival, and Ye Wang, ACM Multimedia 2009. pdf
  • Generating Targeted Rhythmic Exercises for Music Students with Constraint Satisfaction Programming. Graham Percival, Torsten Anders, and George Tzanetakis, ICMC 2008. pdf
  • Effective Use of Multimedia for Computer-Assisted Musical Instrument Tutoring. Graham Percival, Ye Wang, and George Tzanetakis, EMME 2007. pdf
  • Analysis of Saxophone Performance for Computer-Assisted Tutoring. Matthias Robine, Graham Percival, and Mathieu Lagrange, ICMC 2007. pdf
  • Pedagogical Transcription for Multimodal Sitar Performance. Ajay Kapur, Graham Percival, Mathieu Lagrange, and George Tzanetakis, ISMIR 2007. pdf
  • Adaptive Harmonization and Pitch Correction of Polyphonic Audio using Spectral Clustering. Mathieu Lagrange, Graham Percival, and George Tzanetakis, DAFX 2007. pdf

Master’s Thesis: Computer-Assisted Musical Instrument Tutoring with Targeted Exercises

Abstract

Learning to play a musical instrument is a daunting task. Musicians must execute unusual physical movements within very tight tolerances, and must continually adjust their bodies in response to auditory feedback. However, most beginners lack the ability to accurately evaluate their own sound. We therefore turn to computers to analyze the student’s performance. By extracting certain information from the audio, computers can provide accurate and objective feedback to students.

This thesis lays out some general principles for such projects, and introduces tools to help practicing rhythms and violin intonation. There are three distinct portions to this research: automatic exercise creation, audio analysis, and visualization of errors. Exercises were created with Constraint Satisfaction Programming, audio analysis was performed with amplitude and pitch detection, and errors were displayed with a novel graphical interface. This led to the creation of MEAWS, an open-source program for music students.

Oral presentation

At UVic, the thesis defense begins with a 15-20 minute presentation. Here are the slides I used: camit-talk.pdf

Full text

masters-percival.pdf, (approx 1 Mb)