A survey of deep learning audio generation methods

作     者：Božić, Matej Horvat, Marko 

作者机构：Department of Applied Computing University of Zagreb Faculty of Electrical Engineering and Computing Zagreb Croatia 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2024年

核心收录：

主　　题：Deep learning 

摘      要：This article presents a review of typical techniques used in three distinct aspects of deep learning model development for audio generation. In the first part of the article, we provide an explanation of audio representations, beginning with the fundamental audio waveform. We then progress to the frequency domain, with an emphasis on the attributes of human hearing, and finally introduce a relatively recent development. The main part of the article focuses on explaining basic and extended deep learning architecture variants, along with their practical applications in the field of audio generation. The following architectures are addressed: 1) Autoencoders 2) Generative adversarial networks 3) Normalizing flows 4) Transformer networks 5) Diffusion models. Lastly, we will examine four distinct evaluation metrics that are commonly employed in audio generation. This article aims to offer novice readers and beginners in the field a comprehensive understanding of the current state of the art in audio generation methods as well as relevant studies that can be explored for future research. Copyright © 2024, The Authors. All rights reserved.