Discriminative feature based on FWMW for playback speech detection
Discriminative feature based on FWMW for playback speech detection
复制标题
基于FWMW的判别特征用于回放语音检测
DOI：
10.1049/el.2019.1025
复制
发表时间：
2019-06
影响因子：
1.1
通讯作者：
He Qianhua
中科院分区：
文献类型：
--
作者：
Yang Jichen;Liu Leian;He Qianhua
提出了一种用于回放语音检测的判别式特征提取方法，该方法基于逐帧幅度谱加权（FWMW）能扩大真实语音与回放语音之间的差异这一发现。通过将每个帧幅度谱频点的和作为帧幅度谱的权重来获得所建议的FWMW。在此基础上，提出了一种新的基于FWMW的特征，即恒Q权重分割系数（CQWSC）。使用CQWSC在ASVspoof 2017版本2.0评估集上的实验结果表明：（i）FWMW可以使所提出的特征具有更强的区分能力，在高斯混合模型和深度神经网络下的等错误率分别下降15.53%和19.18%，（ii）CQWSC的性能优于一些常用的特征。
A discriminative feature extraction method for playback speech detection is proposed, it relies on the finding that frame-wise magnitude-spectrum weight (FWMW) can enlarge the difference between genuine speech and playback speech. The proposed FWMW is obtained by supplying the sum of every frame magnitude spectrum frequency bin as the weight on the frame magnitude spectrum. Then, a new feature based on FWMW is proposed, namely constant-Q weight segmentation coefficients (CQWSCs). The experimental result on ASVspoof 2017 version 2.0 evaluation set using CQWSC indicates that: (i) FWMW can make the proposed feature has more discriminative ability and the equal error rate can decline 15.53 and 19.18% under Gaussian mixture model and deep neural network, respectively, and (ii) CQWSC performs better than some commonly used features.
影响因子：
2.4
作者：
J. Wolf
通讯作者：
J. Wolf
影响因子：
1.1
作者：
Jichen Yang;Lei-an Liu
通讯作者：
Jichen Yang;Lei-an Liu