About

Marsyas (Music Analysis, Retrieval and Synthesis for Audio Signals) is an open source software framework for audio processing with specific emphasis on Music Information Retrieval applications. It has been designed and written by George Tzanetakis (gtzan@cs.uvic.ca) with help from students and researchers from around the world. Marsyas has been used for a variety of projects in both academia and industry.

Featured Project

Personalized Multimodal Music Search

Marsyas has been helping a lot in our music search prototype called "Personalized Multimodal Music Search", built in Dr Wang Ye's group at School of Computing, National University of Singapore. We mainly use Marsyas for music classification using audio signals. More specifically, we constructed our own classification networks using Marsyas modules for genre, mood, instrument, and vocalness classifications. The class activation probabilities in the classification results were used as audio signatures to represent different music dimensions (namely, genre, mood, instrument and vocalness). The music search prototype is publicly accessible from the link below . Besides searching music by its content, the search engine also provides music search by keywords. In addition, the system allows users to personalize different music dimensions to do their search by keyword or example (only mp3 examples can be recognized for now) . For more details of the system, please refer to the SIGIR'09 paper titled "CompositeMap: A Novel Framework for Music Similarity Measure". We really benefited a lot from Marsyas framework in implementing the audio processing module of our system. We thank all the contributors of Marsyas for their great efforts. - Bingun (Eddie) Zhang, Ye Wang, National University of Singapore

Featured Developer

Coder Steven Ness

Affiliation : University of Victoria

Steven is a 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 like cantillion and pan.sness.net. He's currently doing his M.Sc. in the lab of George Tzanetakis in the field of Music Information Retrieval.

Featured Web Demo

Panning Pedagogy Panning Pedagogy

A Flash application that, using Marsyas, takes audio data, calculates the spectrum for the left and right channels, and calculates the Stereo Panning Spectrum from this data. It is then graphically displayed in a web application. Look for similar programs inside of Marsyas that display this data using 3D OpenGL.

Featured Video

KeyBoard Controlled Assistive Music Browser

The goal of this project is to develop Assistive Music Browser Software that will enable users with severe disabilities to navigate through their own collections of music via a variety of input methods. We extract relevant features and analyze the audio data within a collection of music (of essentially unlimited size) and map the songs onto a 2-D space, called a self-organizing map, based on similarity. Continual audio feedback is provided to the user as they search their collection to aide in navigation and selection. http://www.canassist.ca