Visualization in audio-based music information retrieval

Music information retrieval (MIR) is steadily growing as a research area, as can be evidenced by the international conferences on music information retrieval (ISMIR) series and the increasing number of MIR-related publications. At present, various visualization techniques developed in the context of music information retrieval for representing polyphonic audio signals are already available. The techniques fall into two major categories: techniques for visualizing a single file or piece of music, and techniques for visual collections of pieces. The former includes similariy matrix, beat spectrum and beat spectrogram, beat histograms, real-time audio classification display, and mapping time-varying timbre to color. The latter covers timbre spaces, music similarity via self-organizing maps and smoothed data histograms, and combining different views. Regardless of the category, all of these techniques use sophisticated analysis algorithms to automatically extract content information from music stored in digital audio format.