SPM Special Issue on Recent Advances in Music Signal Processing

Music is a ubiquitous and vital part of our lives. Thanks to the digital revolution in music distribution and storage, music has become one of the most popular categories of multimedia content. In general terms, music processing research contributes concepts, models, and algorithms that extend our capabilities of accessing, analyzing, manipulating, and creating music. Given the complexity and diversity of music, researchers must account for various aspects, such as the genre, instrumentation, dynamics, tempo, rhythm, and timbre. Music signals typically comprise a wide range and large number of different sound sources. Postprocessing and the use of audio effects in the mixing and mastering stages may further complicate the analysis of recorded musical material. Furthermore, music is inherently multimodal, incorporating speech-like signals (singing), video (of live performances), and still images (scanned music scores). This wealth of data makes music processing a challenging field of research and closely connected to areas such as audio and acoustic signal processing, multimedia signal processing, and machine learning.

Compared with speech processing, a research field with a long tradition, music processing is still a relatively young discipline, but it is rapidly growing. In recent decades, the music processing community has come together by organizing major annual conferences on topics including music information retrieval, sound and music computing, audio effects processing, computer music, and applications in audio engineering. Although computer-based music research has traditionally been conducted using symbolic representations, the research focus has shifted to other types of music related data including audio recordings, digitized images, music videos, and other types of sensor data. As a consequence, digital signal processing has found its way into many research communities dealing with music-related data.

In the special issue of IEEE Signal Processing Magazine (SPM) in Jan. 2019, the editors survey recent advances in music processing with a focus on audio signals. Eleven articles cover topics including music analysis, retrieval, source separation, singing-voice processing, musical sound synthesis, and user interfaces, to name a few. The tutorial-style articles provide an overview of theory and applications and discuss main advances. Although music processing has benefited a lot from traditional fields, such as signal processing, the editors hope to convince the reader that the rich and challenging problem domain of music also has many things to offer to signal processing and other research disciplines.