Treffer: Introduction to Differentiable Audio Synthesizer Programming.

Differentiable digital signal processing is a technique in which signal processing algorithms are implemented as differentiable programs used in combination with deep neural networks. The advantages of this methodology include a reduction in model complexity, lower data requirements, and an inherently interpretible intermediate representation. In recent years, differentiable audio synthesizers have been applied to a variety of tasks, including voice and instrument modelling, synthesizer control, pitch estimation, source separation, and parameter estimation. Yet despite the growing popularity of such methods, the implementation of differentiable audio synthesizers remains poorly documented, and the simple formulation of many synthesizers belies their complex optimization behaviour. To address this gap, this tutorial offers an introduction to the fundamentals of differentiable synthesizer programming. The tutorial will centre around practical demonstrations, which participants can follow using an accompanying suite of Jupyter notebooks. All tutorial content will be documented in an accompanying web book, and all tutorial materials and dependencies will be fully open source and accessible for free online. Prior experience with writing Python 3 code is assumed, and a basic knowledge of PyTorch is beneficial though not strictly required. The tutorial is targeted at music and audio researchers and engineers with a grounding in the basics of digital signal processing and machine learning. Our aim is to equip participants to apply these techniques in their own research, whilst enabling those with prior knowledge to sharpen their skills. [ABSTRACT FROM AUTHOR]

Copyright of International Society for Music Information Retrieval Conference Proceedings is the property of Ubiquity Press and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)