Minglei Ma

2 accepted papers

2024

GMM-ResNet2: Ensemble of Group Resnet Networks for Synthetic Speech Detection

Zhenchun Lei, Hui Yan, Changhong Liu, Yong Zhou, Minglei Ma

ICASSP 2024accepted

Deep learning models are widely used for speaker recognition and spoofing speech detection. We propose the GMM-ResNet2 for synthesis speech detection. Compared with the previous GMM-ResNet model, GMM-ResNet2 has four improvements. Firstly, the different order GMMs have different capabilities to form…

Cited by 0 Source ↗Scholar ↗

2022

Two-Path GMM-ResNet and GMM-SENet for ASV Spoofing Detection

Zhenchun Lei, Hui Yang, Changhong Liu, Minglei Ma, Yingen Yang

ICASSP 2022accepted

The automatic speaker verification system is sometimes vulnerable to various spoofing attacks. The 2-class Gaussian Mixture Model classifier for genuine and spoofed speech is usually used as the baseline for spoofing detection. However, the GMM classifier does not separately consider the scores of f…

Cited by 0 Source ↗Scholar ↗