Lecture: Music Processing Analysis, Winter Term 2018/2019

01_MusicRepr_Teaser 02_FourierTr_Teaser 03_MusicSync_Teaser2 04_AudioStru_Teaser 06_Teaser_BeatTempo 07_Teaser_AudioRetr


  • Instructor: Prof. Dr. Meinard Müller
  • Tutor: Frank Zalkow
  • Credits:
    • 2.5 ECTS (Lecture only)
    • 5 ECTS (Lecture with Exercises; only for Computer Science Students)
  • Time (Lecture): Winter Term 2019/2020, Mo 16:15–18:00
  • Time (Exercises): Winter Term 2019/2020, Mo 14:15–16:00
  • Place: Am Wolfsmantel 33, Erlangen-Tennenlohe, Room 3R4.04
  • 1. Lecture: 14.10.2019
  • 1. Excercise: 21.10.2019
  • Exam (graded): Oral examination at the end of term
  • Flyer: PDF
  • Dates (Lecture) (16:15–18:00, Room 3R4.04): Mo 14.10.2019, Mo 21.10.2019, Mo 28.10.2019, Mo 11.11.2019, Mo 18.11.2019, Mo 25.11.2019, Mo 02.12.2019, Mo 09.12.2019, Mo 16.12.2019, Mo 13.01.2020, Mo 20.01.2020, Mo 27.01.2020, Mo 03.02.2020
  • Examination Dates (Room 3R4.03): To be announced

Format

The lecture has the following format:

  • Every meeting consists of 90 minutes.
  • There will be additional exercises for computer science students (see below for details).

For further information, please contact Prof. Dr. Meinard Müller.

Content

Music signals possess specific acoustic and structural characteristics that are not shared by spoken language or audio signals from other domains. In fact, many music analysis tasks only become feasible by exploiting suitable music-specific assumptions. In this course, we study feature design principles that have been applied to music signals to account for the music-specific aspects. In particular, we discuss various musically expressive feature representations that refer to musical dimensions such as harmony, rhythm, timbre, or melody. Furthermore, we highlight the practical and musical relevance of these feature representations in the context of current music analysis and retrieval tasks. Here, our general goal is to show how the development of music-specific signal processing techniques is of fundamental importance for tackling otherwise infeasible music analysis problems.

The lecture closely follows the textbook Fundamentals of Music Processing (FMP). Additionally, the FMP Notebooks offer a collection of educational material, providing detailed textbook-like explanations of central techniques and algorithms in combination with Python code examples that illustrate how to implement the theory.

The following video gives a brief impression about this course.

Course requirements

In this course, we discuss a number of current research problems in music processing or music information retrieval (MIR) covering aspects from information science and digital signal processing. We provide the necessary background information and give numerous motivating examples so that no specialized knowledge is required. However, the students should have a solid mathematical background. The lecture is accompanied by readings from the textbook Fundamentals of Music Processing (FMP) and the research literature. Furthermore, the students are required to experiment with the presented algorithms using Python and/or MATLAB.

Links

Accompanying Textbook and Notebooks

Cover_Mueller_FMP_small

Meinard Müller
Fundamentals of Music Processing
Audio, Analysis, Algorithms, Applications
ISBN: 978-3-319-21944-8
Springer, 2015

FMP Notebooks
Comprehensive framework based on Jupyter notebooks
for teaching and learning fundamentals of music processing.

Exercises (for Computer Science Students)

The exercises, which are particularly provided for computer science students, accompany and extend the lecture Music Processing Analysis. In the exercise meetings, we review the lecture, discuss homework problems, deal with programming issues, and realize mini projects that implement basic algorithms and procedures. If you have any questions regarding the exercise, please contact Frank Zalkow.

Exercise Meeting: 21.10.2019 (14:15–16:00)
  • Organization and announcements
  • Introduction to FMP Notebooks
  • Introduction to Python and Jupyter Notebook
  • Introduction of practical exercises (Due: 18.11.2018)
    • Practical Exercise 1: Short-Time Fourier Transform (Jupyter)
    • Practical Exercise 2: Harmonic-Percussive Source Separation (Jupyter)
    • Each group has to hand in solutions and Python implementations of the practical exercises which are to be presented in the meeting on 18.11.2018.
  • Homework:

Topics

  • Introduction
    Slides (PDF), Handouts (6 slides per page) (PDF)
  • Overview
    Slides (PDF), Handouts (6 slides per page) (PDF)
  • Music Representations
    Slides (PDF), Handouts (6 slides per page) (PDF)
    Literature: Chapter 1
  • Audio Features (Fourier Transform, Spectrogram, Pitch, Chroma)
    Slides (PDF), Handouts (6 slides per page) (PDF)
    Handwritten Notes (Fourier Transform as Optimization Problem) (PDF)
    Literature: Section 2.1, Section 3.1
  • Music Synchronization (Dynamic Time Warping)
    Slides (PDF), Handouts (6 slides per page) (PDF)
    Literature: Section 3.2, Section 3.3
  • Music Structure Analysis
    Slides (PDF), Handouts (6 slides per page) (PDF)
    Literature: Chapter 4
  • Harmony Analysis
    Slides (PDF), Handouts (6 slides per page) (PDF)
    Literature: Chapter 5
  • Tempo and Beat Tracking
    Slides (PDF), Handouts (6 slides per page) (PDF)
    Literature: Chapter 6
  • Audio Retrieval
    Slides (PDF), Handouts (6 slides per page) (PDF)
    Literature: Chapter 7
  • Audio Decomposition
    Slides (PDF), Handouts (6 slides per page) (PDF)