TERASCALE MUSIC MINING

ABSTRACT

The objective of the International Music Information Retrieval Systems Evaluation Laboratory project (IMIRSEL) is the establishment of the necessary resources for the scientifically valid development and evaluation of emerging Music Information Retrieval (MIR) and Music Digital Library (MDL) techniques and technologies. IMIRSEL is located at the Graduate School of Library and Information Science (GSLIS) at the University of Illinois at Urbana- Champaign (UIUC). Project Principal Investigator is J. Stephen Downie of GSLIS and Co-Principal Investigator is Prof. Michael Welge of the Automated Learning Group (ALG) of the National Center for Supercomputing Applications (NCSA). A mission of the project is the creation of secure, yet accessible, terascale collections of music materials in a variety of audio, symbolic and metadata forms. These collections, when coupled with a set of standardized experimental tasks and standardized evaluation metrics, will allow members of the international MIR/ MDL research community to participate in the newly created, TREC-like, Music Information Retrieval Evaluation eXchange (MIREX) contests.

REFERENCES

• 1: [fr1] Small, 100 hour, collections of CD quality audio (16bit sample size, 44.1KHz sample rate)require ~ 60GB of storage. We expect our collection of Naxos audio files (i.e., the CD quality audio from 5000 CDs), when completed, to be in the 4TB range. The Naxos collection is only one of the test collections being used by IMIRSEL.
• 2: [fr2] Some audio processing techniques can run 40-50 longer than real-time.
• 3: J. Stephen Downie , Andreas F. Ehmann , David Tcheng, Real-time genre classification for music digital libraries, Proceedings of the 5th ACM/IEEE-CS joint conference on Digital libraries, June 07-11, 2005, Denver, CO, USA [doi>10.1145/1065385.1065480]
• 4: [fr4] Perrot, D., and R. Gjerdigen. (1999). Scanning the dial: An exploration of factors in identification of musical style. In Proceedings of the Society for Music Perception Cognition, p. 88.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications
          Subjects: Data mining

Additional Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.7 Digital Libraries
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.5 Sound and Music Computing

General Terms:
Design, Experimentation, Measurement, Performance