This website provides the lecture / seminar materials of different lectures on the topic of machine listening.

Please contact Dr.-Ing. Jakob Abeßer @ jakob.abesser@idmt.fraunhofer.de for questions.

Licence

All lecture material is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)

Computational Analysis of Sounds and Music

This website provides the slides & materials for the lecture Computational Analysis of Sound and Music, which will be held during the summer semester 2024 at the TU Ilmenau.

Machine Listening for Music and Sound Analysis

This lecture will be held for the fourth time in the Winter semester 2023/24 as part of the Audio Signal Processing & Audio Systems Module at TU Ilmenau.

Last Update: 07.11.2023

Lecture Material

• Intro - Slides (PDF)
• Lecture 1 - Introduction to Audio Representations - Slides (PDF)
• Lecture 2 - Introduction to Machine Learning - Slides (PDF)
• Lecture 3 - Music Information Retrieval I - Slides (PDF)
• Lecture 4 - Music Information Retrieval II - Slides (PDF)
• Lecture 5 - Environmental Sound Analysis I - Slides (PDF)

• Lecture 6 - Environmental Sound Analysis II - Slides (PDF)

• Audio examples included in the slides can be downloaded as “audio.zip” from Zenodo

Video recording of the lecture

• Please contact Jakob Abeßer to get access to the video recordings of last year’s lectures (except for the first lecture)

Seminar Material

Seminars are held by Dipl.-Ing. Christian Kehling.

• Seminar 1 - Introduction to Python/Audio Processing Basics - IPython Notebook (ipynb), html, Piano1-1.wav
• Seminar 2 - Sound Classification using Support Vector Machine - ipynb, html
• Seminar 3 - Sound/Music Classification with Neural Networks - ipynb, html;

AI-based Audio Analysis (KI-gestützte Audioanalyse von Musik und Soundscapes)

(HfM Weimar, Winter Semester 2022/2023, Dr.-Ing. Jakob Abeßer + Prof. Dr. Martin Pfleiderer)

This seminar will be held jointly with Prof. Martin Pfleiderer (HfM Weimar) in the Winter semester 2022/23 on a bi-weekly basis.

Last Update: 08.11.2022

Lecture Material (Slides / Jupyter Notebooks)

• AIAA 0 - Introduction
  • Slides (PDF)
• AIAA 1 - Python
  • Slides (PDF) - How to install / use python (local, Jupyter Notebook, Google Colab)
    • Config file (YML) for conda environment
  • Slides (PDF) - Python Basics
  • Jupyter Notebook (ipynb)
    • Open in Google Colab
• AIAA 2 - Audio Processing
  • Slides (PDF)
    • Audio Examples (ZIP with WAV files) - need to be unzipped into subfolder called “audio”
  • Jupyter Notebook (ipynb)
    • Open in Google Colab
    • Audio Examples
      • bird.wav
      • piano.wav
• AIAA 3 - Research Projects
  • Slides (PDF)
• AIAA 4 - Machine Learning
  • Slides (PDF)
  • Jupyter Notebook (ipynb)
    • Open in Google Colab
    • Audio Examples
      • animal_sounds.zip - Unzip this file in a subdirectory called audio, it contains 25 WAV files in 5 subdirectories. The folder structure must be /audio/animal_sounds/cats/…
• AIAA 5 - Deep Learning
  • Slides (PDF)
  • Jupyter Notebook (ipynb)
    • Open in Google Colab
    • This notebook again uses the audio examples from the previous lecture (animal classification)
• AIAA 6 - Research Project - Useful hints
  • Additional hints for the research project
    • How to read dataset metadata from CSV files?
    • How to segment audio files into uniform-length segments?
    • Dataset split into training and test set
    • Alternative audio features for specific tasks
    • Research report structure
    • Resources for literatur research
  • HTML
  • Jupyter Notebook (ipynb)