Bibliographic record
Abstract
In recent years, advancements in artificial intelligence have led to a surge in the generation of synthetic and manipulated audio, commonly referred to as "deepfake audio." While these technologies offer advantages across various domains, they also present serious security and ethical concerns, particularly in contexts where the authenticity of audio is critical. This paper introduces a novel deep learning-based approach for detecting deepfake audio using a combination of Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and an Attention mechanism. The proposed architecture utilizes CNNs to extract high-level spatial features from audio spectrograms, while the LSTM network captures the temporal dependencies inherent in audio sequences. The integration of the Attention mechanism further enhances the model's ability to focus on key segments of the audio that are more likely to contain deceptive artifacts. Through comprehensive experimentation on publicly available datasets, our model demonstrates superior performance in terms of accuracy and robustness compared to traditional and standalone deep learning models. These findings underscore the potential of hybrid architectures in effectively addressing the challenges of deepfake audio detection and contribute to the development of trustworthy audio verification systems. KEYWORDS deepfake audio detection, synthetic audio, machine learning, digital forensics, neural networks, feature extraction, deep learning, audio synthesis, data integrity, security
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".