Stochastic gradient descent analysis for the evaluation of a speaker recognition

详细信息    查看全文

• 作者：Ashrf Nasef ; Marina Marjanović-Jakovljević…
• 关键词：Pattern recognition ; Speech analysis ; Deep learning Neural Network ; Stochastic gradient descent ; Learning rate ; Dropout rate
• 刊名：Analog Integrated Circuits and Signal Processing
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：90
• 期：2
• 页码：389-397
• 全文大小：
• 刊物类别：Engineering
• 刊物主题：Circuits and Systems; Electrical Engineering; Signal,Image and Speech Processing;
• 出版者：Springer US
• ISSN：1573-1979
• 卷排序：90

文摘

Performance optimization in speaker recognition is a challenging task in the field of vocal based human-computer interaction. Many researches have shown that deep learning Neural Network methods have the best performance in comparison with other classifiers. However, those methods with many parameters require a lot of tunings in order to optimize the performance in different supervised learning tasks. In this paper, we show that picking a good combination of parameters can significantly improve the performance of Stochastic Gradient Descent deep learning Neural Network method in automatic speaker recognition even in a noisy environment. Parameters that are analyzed are learning rate, hidden and input layer dropout rate.