Noise-robust speaker verification using F0 features
Noise-robust speaker verification using F0 features
复制标题
使用 F0 功能进行抗噪声扬声器验证
DOI：
10.21437/interspeech.2004-503
复制
发表时间：
2004
期刊：
影响因子：
--
通讯作者：
S. Furui
中科院分区：
文献类型：
--
作者：
K. Iwano;Taichi Asami;S. Furui
本文提出了一种基于基频（F0）增强的抗噪声说话人确认方法。本文首先描述了一种使用Hough变换的噪声鲁棒F0提取方法。在此基础上，提出了一种基于F0和倒谱特征融合的多流HHT说话人确认方法。实验是在半年的时间里，由37名男性在五个会话中记录的日语四位连数字话语进行的。在不同的SNR水平下，话语被白色噪声污染。实验结果表明，F0特征在所有信噪比条件下都能提高验证性能。
This paper proposes a noise-robust speaker verification method augmented by fundamental frequency (F0). The paper first describes a noise-robust F0 extraction method using the Hough transform. Then, it proposes a robust speaker verification method using multi-stream HMMs which fuse the extracted F0 and cepstral features. Experiments are conducted using fourconnected-digit utterances of Japanese by 37 male speakers recorded at five sessions over a half year period. The utterances are contaminated with white noise at various SNR levels. Experimental results show that the F0 features improve the verification performance in all SNR conditions.