Instructor: Martin Rocamora.

This course provides a survey of the field of Music Information Retrieval (MIR). Broadly speaking, the field of MIR is foremost concerned with the extraction and inference of meaningful features from music (from the audio signal, symbolic representation or external sources such as web pages), indexing of music using these features, and the development of different search and retrieval schemes (for instance, content-based search, music recommendation systems, or user interfaces for browsing large music collections). As a consequence, MIR aims at making the world’s vast store of music available to individuals.

In this course a special emphasis is given to well-established techniques for the automatic description of audio content in terms of different facets (e.g. melody, harmony, rhythm, timbre), temporal scopes, and abstraction levels (from low-level features to semantic descriptions such as genre or mood). In addition, we review methods for feature extraction and music indexing from contextual data sources about music items, such as web pages or collaborative tags. The combination of music content and context description enables a wide variety of music retrieval tasks, such as semantic music search or music identification. We examine how various MIR approaches to different problems are evaluated and discuss the major open challenges in the field.

Content

  1. Introduction to MIR: history and evolution, music modalities and representations, applications, research topics and tasks.
  2. Levels and facets of music content description.
  3. Low-level descriptors and timbre.
  4. Pitch content description.
  5. Novelty detection and segmentation.
  6. Chroma, chords and key estimation.
  7. Rhythmic analysis.
  8. Music similarity.
  9. Classification and auto-tagging.
  10. Context-based music description and indexing.
  11. User properties and user context.
  12. Evaluation in Music Information Retrieval.
  13. Open challenges.

Format

The course is offered in 10 weeks during the 2nd Term (January-March), with 25 hours of lectures.

Class format: 2.5 h/week including lectures, hands-on exercises, and discussions.

Each week, all students are expected to review the lecture material and to work on a set of questions proposed at the lectures and practical assignments proposed at the hands-on sessions (labs).

The evaluation of the course is based on the following items: lab reports (60%), final exam (30%) and participation in class (10%)

Materials and References

https://github.com/MTG/MIRCourse