George Tzanetakis

Main Designer and Developer

Affiliation: University of Victoria - Department of Computer Science


I am an assistant professor in Computer Science (also cross-listed in Music and Electrical and Computer Engineering) at the University of Victoria in Canada. I received my PhD in Computer Science from Princeton University under the supervision of Dr. Perry Cook. I also worked at Carnegie Mellon University as a PostDoctoral Fellow with Dr. Roger Dannenberg on query-by-humming systems and audio-score alignment and with the Informedia group on multimodal video retrieval and microphone arrays. I have also consulted with several companies using Marsyas. They include Moodlogic Inc. (audio fingerpriting), All Music Inc., The Netherlands (music-speech classification), and Teligence Communications Inc. (gender classification of voice recordings). My research deals with all stages of audio content analysis such as feature extraction, segmentation, classification, retrieval and source separation with specific focus on Music Information Retrieval (MIR). I am also an active musician and have studied saxophone performance, music theory and composition.

Graham Percival

Lilypond, Documentation Editor


I finished my Masters in Computer Science and Music, working on software to help beginning string students. I am also the Documentation Editor and Bug Meister for GNU/LilyPond.

Jakob Leben


I am a researcher with main interest in models of computation and programming languages for digital signal processing.

My contributions to Marsyas include:

  • Marsyas Script (scripting language)
  • Marsyas Inspector (graphical network inspector),
  • improved real-time audio processing
  • improved Open Sound Control support and real-time control in general
  • improved interaction with Qt
  • dataflow testing and debugging system

I have also contributed significantly to SuperCollider (programming language and system for sound synthesis and algorithmic composition) and other open-source projects.

I am currently a PhD student at Computer Science, Universtiy of Victoria under supervision of prof. George Tzanetakis. My current research focuses on the design of a new programming language for DSP and compiler optimization techniques.

Luis Gustavo Martins

Windows and Qt Guru (Portuguese Invasion 1)


I've been using and contributing to Marsyas since its 0.1 version, back in 2003, when working on a contract project that required a realtime audio classifier (for music/speech/noise/silence...). When I started my PhD in the area of computational auditory scene analysis (CASA), I based all my research software implementation and evaluations in Marsyas 0.2. Marsyas was a key factor for my PhD work, being efficient and flexible to develop advanced algorithms for sound segregation, that run close to real-time. Contributing to Marsyas also allowed me to interact and work with all the Marsyas developer team, becoming one of the most import learning experiences in my research and academic life.

Steven Ness



Steven is a mad scientist and coder who works on various aspects of Marsyas. He added unit tests to Marsyas, turned the website into a Ruby on Rails backend, has written some Marsystems, and uses Marsyas in the development of rich internet-enabled applications. He's currently doing his Ph.D. in the lab of George Tzanetakis in the field of Music Information Retrieval.

Mathieu Lagrange

Author of the most complex MarSystem network


I am interested in auditory scene analysis for music streams understanding. For such a purpose, Marsyas as a valuable tool as it provides the user with an efficient, easy-to-install platform for processing various type of audio data. As the owner of the Most complex Marsystem network Award, I can say that even on challenging research tasks, Marsyas proved to be flexible and fun to use.

Stefaan Lippens

Coder and Code Styler


My first encounter with Marsyas was in 2002. It was used for my master thesis at Ghent University on the topic of automatic musical genre classification. After completing my electrical engineering studies, I started working in the field of image processing and obtained my PhD on halftoning and printing. Since 2008 I'm back in the field of audio and music processing, now as a post-doctoral researcher at the Digital Speech and Signal Processing research group of Ghent University. The research revolves around music information retrieval and is bridged with an outside company. Together we focus on large scale automatic extraction of several music characteristics such as musical genre and rhythm style. My main Marsyas activities and contributions are situated in the core MarSystems and the Python bindings.

Tiago F. Tavares


I'm a PhD candidate in the University of Campinas (UNICAMP) and I will be visiting UVic, here in Victoria-BC, for one year. I have been working with automatic transcription of audio for some years now, and I hope to contribute with the development of Marsyas as much as I can. I have written some documentation, and let's see what patches I will do in the future!

Adam Tindale

Open Sound Control, Chuck, Percussion and Live Electroacoustic Music


Adam Tindale is an electronic drummer, teacher, and researcher. His research combines signal processing and machine learning tools from Marsyas to classify drum events in real-time to develop a more expressive electronic drum. Adam is currently completing his Interdisciplinary Ph.D. in Music, Computer Science, and Electrical Engineering under the supervision of George Tzanetakis.>

Thijs Koerselman

Software Developer and Designer


I'm a software developer and designer working with interactive media and sound. I hold an MA and BSc in Music Technology. After graduating in 2004 I got increasingly involved with programming. I have developed software for creative applications, live performance systems and art installations. Currently I work for the Utrecht School of Arts in the Netherlands, faculty of Art, Media and Technlogy, where we employ Marsyas in a project focusing on flexible and intelligent media repository software. Currently Marsyas is used for tasks such as music/speech classification and similarity matching. All content processing is done via a modular distributed pipeline framework, so additional algorithms can be easily plugged in. Other parts of the project include video analysis, data modeling and adaptive user interfaces.

Luis Teixeira

Video, Python, upcoming Marsyas-0.x (Portuguese Invasion 2)


I'm a PhD student at FEUP and a researcher at the Telecommunications and Multimedia Unit of INESC Porto. Currently most of my time is consumed by my PhD and by the strange experiments I'm doing with Marsyas like trying to get video to work in it, and who knows what more! As for the PhD, the focus is on the detection of events and automatic description of multi-sensor systems. Previously I worked with MPEG-4 and MPEG-7 for a video editing framework during MSc. That was back in 2004. Before, i.e. since I started my collaboration with INESC Porto in 2001 until 2004, I collaborated in several research projects mainly on distributed multimedia systems. Multimodal analysis, fusion of information from multiple types of sources and multimedia distributed systems are my main research interests. C/C++ and Python are the tools of the trade.

Fabien Gouyon


I have been using Marsyas since 2007 for almost all audio processing involved in my research and that of my students.

Zhang Bingjun (Eddy)



I am currently PhD candidate under the supervision of Dr Wang Ye, in Department of Computer Science, School of Computing, National University of Singapore. My research interest include music information retrieval, multimodal data fusion, and machine learning. In the project of multimodal music information retrieval, we employed Marsyas to build a music analysis module. In addition, we also modified parts of the Marsyas framework to extend it functionality and robustness.

Miguel Lopez


My name is Miguel Lopes, I'm a finalist student at FEUP (Porto), and I've just finished my Masters Degree Thesis about musical genre classification - developed at INESC Porto (Fabien Gouyon was my thesis advisor). I used Marsyas to extract features from audio files and to run several classification experiments using Weka. My thesis consists on classification experiments on the Latin Music Database (presented by Silla, Koerich and Kaestner). A performance comparison between various Weka classifiers and Gaussian Mixture Models is made; there is an assessment of the influence on the classification results of the use of an artist filter, the size of the datasets used and the testing method (cross validation vs different percentages split); there is a comparison between song classification and frame classification. A detailed analysis of the LMD genres and how well each of them is defined in the context of the LMD was also made. Marsyas was used to extract the features from the LMD audio samples (using bextract).

Ajay Kapur

Sensors and Robots


I have been using Marsyas to do audio feature extraction and machine learning experiments in my research in computational ethnomusicology. I have also used Marsyas in live electronic music performance, integrating multimodal sensor interfaces with custom built robotic systems. Director of Music Technology at California Institute of the Arts Professor in Sonic Arts, New Zealand School of Music

Mark Brand


I am lecturer in music technology at the Nelson Mandela Metropolitan University (South Africa), and currently working toward an MScEng from Stellenbosch University under the supervision of Prof. Thomas Niesler (DSP/engineering) and Mr. Theo Herbst (new music). I am investigating, within the MIR domain, alternative music theory approaches in respect of non-western musics, particularly those found in southern Africa. I have a strong bias against the use of common music notation-based theory in this regard, and I'm thus leveraging Marsyas (with much guidance from my supervisors) in a bid to unmask an alternate theoretical framework. Before that I was a rock musician.

Gabrielle Odowichuk


I am pursuing a MASc under the supervision of George Tzanetakis and Peter Driessen at the University of Victoria. My work is in the field of audio signal processing, and I used Marsyas to process real-time audio signals for sound localization using a microphone array. I've written my very own MarSystem to perform cross-correlation, and will use Marsyas for many more projects in years to come. Yay, Marsyas!

Giovanni Donati


I'm an Electronics and Telecommunications Engineering student at the Bologna University in Italy. I'm writing my thesis about Automatic Genre Recognition and Tagging for Music Social Networks. At the moment I'm also incumbent of a scholarship and I'm collaborating with an informatic company called PuzzleDev ( to develope a system called MX-Ray. Basically will be a signal processing based features extractor conceived for web automatic tagging applications, but the final target will be to integrate it into different systems for different purposes. I'm using and will use Marsyas for all the audio processing operations for the prototype because I find it very useful and powerful.

Fabiano Fidancio


I'm a Brazilian software developer/free software enthusiast that found

Aaron Rush


I am a grade 12 student at a high school in Canada. I am interested in the process of transcribing polyphonic music. For such a purpose, Marsyas is a valuable tool as it already has built in features that can be extended to further advance research in this area.