基于深度循环网络的声纹识别方法研究及应用

声纹识别是当前热门的生物特征识别技术之一，能够通过说话人的语音识别其身份。针对声纹识别技术进行了研究，提出了一种基于卷积神经网络（CNN）和深度循环网络（RNN）的声纹识别方案CDRNN，CDRNN结合CNN和RNN的优势，用于移动终端声纹识别应用。CDRNN将说话者的原始语音信息经过一系列的处理并生成一张二维语谱图，利用CNN长于处理图像的优势从语谱图中提取语音信号的个性特征，这些个性特征再输入到Deep RNN中完成声纹识别，从而确定说话者的身份。实验结果表明了CDRNN方案能够获得比GMM-UBM等其他方案更好的识别准确率。

Research and application of deep recurrent neural networks based voiceprint recognition

Voiceprint recognition was one of the most popular biometric identification technologies, which could identify a speaker based on his voice. This paper proposed CDRNN, a voiceprint recognition scheme. CDRNN combined CNN and Deep RNN into a unified model and took advantages of both of them. For CNN was good at extracting characteristics from images, it could generate several spectrograms based on the original voice signal at first. And then, CNN would extract unique features from these spectrograms. . Finally, Deep RNN would output the speaker''s identification based on these unique features. Simulation results show that CDRNN performs better than GMM-UBM and DNN-based approach.