Detail výsledku
Fast Temporal Convolutions for Real-Time Audio Signal Processing
Schimmel Jiří, doc. Ing., Ph.D., UTKO (FEKT)
This paper introduces the possibilities of optimizing neural network convolutional layers for modeling nonlinear audio systems and effects. Enhanced methods for real-time dilated convolutions are presented to achieve faster signal processing times than in previous work. Due to the improved implementation of convolutional layers, a significant decrease in computational requirements was observed and validated on different configurations of single layers with dilated convolutions and WaveNet-style feedforward neural network models. In most cases, equivalent signal processing times were achieved to those using recurrent neural networks with Long Short-Term Memory units and Gated Recurrent Units, which are considered state-of-the-art in the field of black-box virtual analog modeling
convolutional neural networks; deep learning; virtual analog modelling; nonlinear systems
@inproceedings{BUT178795,
author="Štěpán {Miklánek} and Jiří {Schimmel}",
title="Fast Temporal Convolutions for Real-Time Audio Signal Processing",
booktitle="Proceedings of the 25th International Conference on Digital Audio Effects (DAFx20in22)",
year="2022",
journal="Proceedings of the International Conference on Digital Audio Effects (DAFx)",
pages="115--121",
publisher="DAFx",
address="Vídeň",
isbn="978-3-200-08599-2",
issn="2413-6689"
}