Audio deepfakes: A survey

Abstract

A deepfake is content or material that is synthetically generated or manipulated using artificial intelligence (AI) methods, to be passed off as real and can include audio, video, image, and text synthesis. The key difference between manual editing and deepfakes is that deepfakes are AI generated or AI manipulated and closely resemble authentic artifacts. In some cases, deepfakes can be fabricated using AI-generated content in its entirety. Deepfakes have started to have a major impact on society with more generation mechanisms emerging everyday. This article makes a contribution in understanding the landscape of deepfakes, and their detection and generation methods. We evaluate various categories of deepfakes especially in audio. The purpose of this survey is to provide readers with a deeper understanding of (1) different deepfake categories; (2) how they could be created and detected; (3) more specifically, how audio deepfakes are created and detected in more detail, which is the main focus of this paper. We found that generative adversarial networks (GANs), convolutional neural networks (CNNs), and deep neural networks (DNNs) are common ways of creating and detecting deepfakes. In our evaluation of over 150 methods, we found that the majority of the focus is on video deepfakes, and, in particular, the generation of video deepfakes. We found that for text deepfakes, there are more generation methods but very few robust methods for detection, including fake news detection, which has become a controversial area of research because of the potential heavy overlaps with human generation of fake content. Our study reveals a clear need to research audio deepfakes and particularly detection of audio deepfakes. This survey has been conducted with a different perspective, compared to existing survey papers that mostly focus on just video and image deepfakes. This survey mainly focuses on audio deepfakes that are overlooked in most of the existing surveys. This article's most important contribution is to critically analyze and provide a unique source of audio deepfake research, mostly ranging from 2016 to 2021. To the best of our knowledge, this is the first survey focusing on audio deepfakes generation and detection in English.

Open Source and Proprietary Software for Audio Deepfakes and Voice Cloning: Growth Areas, Pain Points, Future Influence

“…Challenges and threats posed by audio deepfakes. While voice cloning and audio deepfake technologies hold significant promise for a range of professional applications, from automated news reporting to personalized content creation, they are beset with challenges that span ethical considerations, computational demands, linguistic diversity, and the intricacies of human speech (Khanjani et al, 2023).…”

“…TTS systems often struggle with homographs-words spelled the same but with different meanings-leading to incorrect pronunciations in context. Furthermore, recognizing periods, special characters, and the nuances of human speech elements like breathing, laughter, and pauses remains challenging, deviating from synthesized speech's human-like quality (Khanjani et al, 2023).…”

Audio Deepfake Approaches

“…Deepfake refers to synthetic information or materials that have been developed or altered using artificial intelligence (AI) technologies, and are intended to be considered authentic. These may include audio, video, picture, and text synthesis [1].…”

“…Firstly, there's Faceswap, widely popularized by Snapchat, allowing users to modify facial features in photographs for playful transformations . [1]. Secondly, Synthesis techniques, powered by generative adversarial networks (GANs), have revolutionized image creation, with models like NVIDIA 112 generating countless variations of images [26].…”