NEMA - Networked Environment for Music Analysis

Networked Environment for Music Analysis (NEMA), Phase I
(Start Date: 1 January 2008)

URL: http://nema.lis.uiuc.edu (being set up)

Project Goals: NEMA brings together the collective projects and the associated tools of six world leaders in the domains of music information retrieval (MIR), computational musicology (CM) and e-humanities research. The NEMA team aims to create an open and extensible webservice-based resource framework that facilitates the integration of music data and analytic/evaluative tools that can be used by the global MIR and CM research and education communities on a basis independent of time or location. To help achieve this goal, the NEMA team will be working co-operatively with the UIUC-based, Mellon-funded, Software Environment for the Advancement of Scholarly Research (SEASR) project to exploit SEASR’s expertise and technologies in the domains of data mining and webservice-based resource framework development.

Leading Institutions:  (PI) Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign; Champaign, Illinois, (PI) J. Stephen Downie, .

(Co-PI) Schulich School of Music, McGill University, Montreal, Quebec, Ichiro Fujinaga,

Key Research Partners

David De Roure (University of Southampton, UK)

Mark Sandler (Queen Mary, University of London, UK)

Tim Crawford (Goldsmiths, University of London, UK)

David Bainbridge (University of Waikato, NZ)

Project Highlights: The proof-of-concept porting of the standalone Music-to-Knowledge (M2K) toolkit to the SEASR web-service environment.

Milestones and Deliverables:
P1: Y0 to end Y0.5 (January 2008 through June 2008)

1. Final NEMA agreements signed (All labs)
2. Lab workers hired (All labs)
3. NEMA All-Hands Meeting I held (All labs + SEASR; Venue to be determined)
4. NEMA Portal server up, including base Greenstone install (UIUC/Waikato)
5. Communication channels established (via NEMA Portal, UIUC)
6. Inter-technology function scoping and prioritizing begun (All labs + SEASR)

P2: Y0.5 to end Y1.0 (July 2008 through December 2008)

1. MIREX DIY (Music Information Retrieval Evaluation eXchange: Do-It-Yourself) vesrion1 up (UIUC + SEASR)
2. NEMA All-Hands Meeting II held (All labs + SEASR, Philadelphia [ISMIR (International Society for Music Information Retrieval) Conference 2008])
3. NEMA Workshop I held (All labs + SEASR, Philadelphia)
4. NEMA servers interconnected (All labs via NEMA Portal @ UIUC)
5. Inter-technology function scoping and prioritizing completed (All labs with UIUC/McGill leading and input from SEASR)
6. Installation of High Level Services on NEMA Portal begun (All labs + SEASR)
7. First tests of the OMEN (On-demand Metadata Extraction Network) model in the NEMA/SEASR environment begun (McGill/UIUC/Waikato first target databases)
8. First tests of Greenstone (Waikato), Maestro (Southampton) and Music Ontology (Queen Mary/Goldsmiths) in the NEMA/SEASR environment begun
9. First modeling of standards for data interchange, interoperability, APIs, etc. begun (All labs + SEASR)
10. Identification of potential early-adopters/trusted users for NEMA Portal begun (All labs)

Community: The NEMA Portal and MIREX DIY play special roles in the Phase I workplan. These two aspects of the NEMA project will function as primary “rallying points” in our campaign to make NEMA a reality. The NEMA Portal, while ultimately intended to be a community-wide resource, will function in its early days as the mechanism through which the NEMA team itself communicates. A first priority of Phase I will be to get the NEMA Portal server up and running. The team will then use the NEMA Portal’s Wikis, Code Repositories, Mail Archives, etc. in an in-house mode to co-ordinate its development efforts. As time goes by, and parts of the NEMA Portal are ready for exposure to the world, these systems will also be opened up to the community. One particular advantage to using the NEMA Portal as an internal NEMA team resource is that it will have had many thousands of hours of use (and debugging) before external users come to use it.

Because MIREX is held each year and has specific deadlines, work on realizing and incrementally improving, the MIREX DIY system is particularly important to the timely development of the NEMA infrastructure. In fact, it is a goal of NEMA to have at least one fully functional, SEASR-based, MIREX DIY prototype system in place by mid-Year 1 so that it can be used in one (or more) of the MIREX 2008 tasks. Also, it will be the MIREX DIY system through which the MIR/CM community will initially become aware of the NEMA vision. This means that the NEMA team will have its first round of real-world user feedback before the end of Year 1 upon which to base developmental improvements. In successive years, the MIREX DIY system will broaden its range of tasks and capabilities and thus will also broaden its user pool from which more useful input can be obtained.

Sustainability: The MIR community has a long-term interest in sustaining MIREX as a crucial tool for comparing various techniques used in music analysis. By making MIREX “Do-It-Yourself” based on open-source software and allowing access to a large amount of testing data, NEMA will naturally be supported by the community. Before Fall 2008 we will test the feasibility of NEMA in the context of MIREX 2008.

Marketing/Evangelism: The NEMA team is multidisciplinary and as such has a collective history of publishing in a variety of venues. These venues include ISMIR, JCDL, Computing in Musicology, ACM Special Interest Group on Information Retrieval, Journal of the American Society of Information Science and Technology, Information Processing and Management, Journal of New Music Research, IEEE Transactions on Multimedia, IEEE Transactions on Audio Speech and Language Processing, IEEE Multimedia Magazine, IEEE Signal Processing Magazine, Digital Humanities, DLib Magazine, Global Grid Forum, and so on. NEMA team members will use these venues as their scholarly communication channels to make sure that NEMA is known to as broad an audience as possible. We also intend to hold at least one NEMA workshop each year with ISMIR being our first-choice venue. NEMA has already been given tentative approval for the first NEMA workshop at ISMIR 2008 in Philadelphia.

Synergy with other projects: The close integration of Mellon-funded SEASR project has already begun and aggregation with other open-source projects such as Greenstone, OMEN, jMIR, OMRAS2, and Maestro, is underway. We will also be looking into a potential collaboration with the Music Information Retrieval Group at Vienna University of Technology led by Andreas Rauber.