◐ Shell
reader mode source ↗
Skip to main page content
Advanced
User Guide
Full text links Cite
Display options

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.

Keywords: artificial intelligence; audio deepfake; cybersecurity; deepfake detection; deepfake generation; misinformation; spoof detection; spoofed audio.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
(A) Audio deepfake generation, text-to-speech. (B) Audio deepfake generation, voice conversion. (C) Audio fake generation frameworks.
Figure 2
Figure 2
Audio deepfake detection frameworks.
Figure 3
Figure 3
Replay attack end-to-end detection. The audio input comes to group delay grams which are novel time-frequency representations of an utterance. Also, the novel attention mechanism softly weights the GD grams. The ResNet-18 network and its GAP layer are used to provide attention maps for a second stage of discriminative training.
Figure 4
Figure 4
Char2Wav. It gets text as input, then brings it to an encoder and then decoder that both of them are RNN based, the output of this phase is linguistic features. The vocoder takes the linguistic features and gives audio.
Figure 5
Figure 5
WaveNet. The input text is presented to causal convolutions, then the output comes to dilated convolutional layers, and then goes to gated activation units, and the activation functions. The Activation Function box includes two (Relu activation functions followed by 1 × 1 layers) plus a Softmax activation function. The figure above also shows the residual block used in WaveNet.
Figure 6
Figure 6
WaveGlow. The text input goes to a single network which is CNN based, and also tries to maximize the likelihood of the training data, and produces the audio output. X is a group of 8 audio samples squeezed as vectors.
Figure 7
Figure 7
(A) Tacotron. The model takes input characters, then goes to some RNN based networks as well as a 1-D convolution bank + highway network + bidirectional GRU. Then, it gives the corresponding raw spectrogram as output, that is fed to the Griffin-Lim reconstruction algorithm for speech synthesizing. (B) Tacotron 2. Text characters are input then they go to embedding, PreNet, three convolutional layers, bidirectional LSTM, attention, 2 LSTM layers (recurrent sequence-to-sequence feature prediction network), Linear Projection, Modified WaveNet as vocoder. Then it outputs the wave. (C) Deep Voice3. Text characters are inputs, then they go for embedding, prenet, convolutional blocks and postnet that all of them are considered as the encoder. The output of this phase goes to the decoder which contains: prenet, attention blocks, causal convolutions, a fully-connected layer, and a binary final frame prediction. Then, it can use one of the existing vocoders for producing audio (WORLD, Griffin-lim and Wavenet).
Figure 8
Figure 8
MelNet. The model gives the input to the three aforementioned stacks. The stacks extract features from different input sections to collectively summarize the entire context.
Figure 9
Figure 9
Impersonation using GAN. The GAN contains 6-layer CNN encoder and transposed6-layer CNN as its generative networks.The discriminative network contains 7-layer CNN with adaptive pooling.
Figure 10
Figure 10
DeepSonar architecture for detecting AI-synthesized fake voices.
All figures (10)
See this image and copyright information in PMC

References

    1. Abe M., Nakamura S., Shikano K., Kuwabara H. (1990). Voice conversion through vector quantization. J. Acoust. Soc. Jpn. 11, 71–76. 10.1250/ast.11.71 - DOI
    1. Ajder H., Patrini G., Cavalli F., Cullen L. (2019). The State of Deepfakes: Landscape, Threats, and Impact, Deeptrace.
    1. Akanksh B., Vekkot S., Tripathi S. (2016). “Interconversion of emotions in speech using td-psola,” in Advances in Signal Processing and Intelligent Recognition Systems (Cham: Springer; ), 367–378. 10.1007/978-3-319-28658-7_32 - DOI
    1. Almutairi Z., Elgibreen H. (2022). A review of modern audio deepfake detection methods: challenges and future directions. Algorithms 15, 155. 10.3390/a15050155 - DOI
    1. Alzantot M., Wang Z., Srivastava M. B. (2019). Deep residual neural networks for audio spoofing detection. arXiv:1907.00501 [cs]. 10.21437/Interspeech.2019-3174 - DOI
Show all 155 references

LinkOut - more resources