Tutorial Scope#
We assume that readers have some prior knowledge in:
Python and NumPy programming (familiarity with Pytorch is a bonus, but not essential)
Spectrogram visualization
Computer-generated sounds
No prior knowledge of wavelet or scattering transforms is expected.
From a theoretical perspective, this tutorial will give an introduction to the use of wavelets in scattering transform. We give practical examples of scattering transforms for audio signals, but mainly those relevant to music analysis.
We aim for this book to be relevant to MIR researchers working on music classification, timbre representation, music similarity, and differentiable signal processing for music analysis and synthesis/digital audio effects.
Tip
You can find more examples in the Kymatio documentation. Join and contribute to the Kymatio community on Github by submitting issues, pull requests, and feature requests. Stay up-to-date with new Kymatio features and releases on X.