# Replay attack detection with complementary high-resolution information using end-to-end DNN for the ASVspoof 2019 Challenge

arXiv: https://arxiv.org/abs/1904.10134 · HTML: https://mlsanity.com/paper/1904.10134/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Apr 2019.

## Built on by (most cited first, 5 of 5)

- [AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks](https://mlsanity.com/paper/2110.01200/) (2021)
- [RW-Resnet: A Novel Speech Anti-Spoofing Model Using Raw Waveform](https://mlsanity.com/paper/2108.05684/) (2021)
- [Phase-Aware Spoof Speech Detection Based on Res2Net with Phase Network](https://mlsanity.com/paper/2203.10793/) (2022)
- [Spoofing-Aware Speaker Verification by Multi-Level Fusion](https://mlsanity.com/paper/2203.15377/) (2022)
- [Beyond Silence: Bias Analysis through Loss and Asymmetric Approach in Audio Anti-Spoofing](https://mlsanity.com/paper/2406.17246/) (2024)

## Abstract (arXiv)

In this study, we concentrate on replacing the process of extracting hand-crafted acoustic feature with end-to-end DNN using complementary high-resolution spectrograms. As a result of advance in audio devices, typical characteristics of a replayed speech based on conventional knowledge alter or diminish in unknown replay configurations. Thus, it has become increasingly difficult to detect spoofed speech with a conventional knowledge-based approach. To detect unrevealed characteristics that reside in a replayed speech, we directly input spectrograms into an end-to-end DNN without knowledge-based intervention. Explorations dealt in this study that differentiates from existing spectrogram-based systems are twofold: complementary information and high-resolution. Spectrograms with different information are explored, and it is shown that additional information such as the phase information can be complementary. High-resolution spectrograms are employed with the assumption that the difference between a bona-fide and a replayed speech exists in the details. Additionally, to verify whether other features are complementary to spectrograms, we also examine raw waveform and an i-vector based system. Experiments conducted on the ASVspoof 2019 physical access challenge show promising results, where t-DCF and equal error rates are 0.0570 and 2.45 % for the evaluation set, respectively.