Data-driven music understanding

"Data-driven music understanding" is an NSF-funded project at LabROSA concerned with extracting information from music audio and discovering deeper patterns and structure within it. This web page is a central hub for materials resulting from this project.

Project reports

• Project Summary and Description from the original NSF proposal
• 2007-8 annual report

Publications

Web materials

• MEAPsoft - software for analyzing and reordering music audio
• Speech and Audio Processing and Recognition:: Class materials (slides, assignments, practicals, demonstrations).
• Automatic piano transcription.
• A baseline artist identification system for the artist20 data in Matlab.
• Artist identification using timbral and chroma features (data and code).
• Automatic chord recognition from music audio (data and code).
• Matlab examples of common audio processing algorithms continually updated.

Acknowledgment

This material is based upon work supported by the National Science Foundation under Grant No. IIS-0713334. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF).

Last updated: $Date: 2008/11/03 19:00:27 $
Dan Ellis <dpwe@ee.columbia.edu>