If you are looking for a sound textbook in music analysis and music information retrieval, I would recommend this one. The author has indeed taken great care in preparing the text for students, teachers, and researchers of music. The book consists of eight chapters, the first two being fundamental.

Chapter 1 addresses the problem of music representations and describes three acknowledged ways of doing it: sheet music, symbolic representations, and audio representations. This is also a chapter that acquaints the reader with terms of music and acoustics such as pitch, timbre, and dynamics. The advanced technical terms, such as Fourier transform and fast Fourier transform (FFT), are, however, reserved for chapter 2. Actual music processing begins in chapter 3 with music synchronization, where the primary task is to “temporally align compatible representations of the same piece of music.” The focus of chapter 4 is on structural analysis of music, while chapter 5 covers harmonic analysis. Tempo and beat analysis are covered in chapter 6, while chapter 7 targets audio retrieval techniques. The final chapter is on audio decomposition, where the author has split the main problem into three subproblems: harmonic-percussive separation, main melody extraction, and score-informed audio decomposition.

The book may be used as a textbook at the postgraduate or advanced undergraduate levels in computational music science. It is also suitable for musicologists and researchers in music. Those with engineering backgrounds will find it easier to grasp.

I enjoyed reading this book. On the plus side, it gives a nice blend of theory and practice in music processing. There are several examples, figures, and exercises. On the minus side, I don’t appreciate the author’s idea of completely skipping analyzing extempore music such as Indian classical music, which has always been an acknowledged challenge in music analysis. “The terms are even more problematic for Indian classical music, where a raga only yields the tonal or melodic framework on which a composition or improvisation is based, rather than denoting a specific piece of music,” writes Müller (p. 385).

Perhaps the author is not aware of the recent book released by the same publisher, which teaches computational musicology in the context of Hindustani (North Indian) music [1]. In fact, Müller’s book is meant only for music with fixed scores (musical notation). Nevertheless, for the overall exposition, Müller deserves praise. Being a statistician cum music analyst myself, I would definitely like to see Müller’s book on the table and not on the shelf!