Sound and music content processing
Description:
The amount and availability of different media (professionally produced or user-generated content) is continuously increasing. This is profoundly changing the ways we interact with sound and music today and how we expect to do it tomorrow. Music Information Retrieval (MIR) is a fast-paced multidisciplinary field of research where different long-tradition disciplines (such as e.g. Signal Processing, Information Science, Computer Science, Musicology) meet to empower these changes. This lecture will provide an overview of MIR research, we will look into ways to extract semantic information from a number of media (from specific audio signals to contextual annotations). A special focus will be put on surveying state-of-the-art algorithms and available systems for the automatic description of music audio signals from a musical perspective (focusing on musically-meaningful dimensions, such as e.g. rhythm, harmony, timbre). We will explore diverse uses of this step of information extraction, such as music search and recommendation, music signal transformations or real-time music performance.
Details:
- Lecturer: Fabien Gouyon.
- Schedule: July 18, 19, 20; 12:00 – 13:30.
- Room: 52s27.
Materials:
