Multidimensional speaker information recognition based on proposed baseline system

Traditional speech-related identity recognition commonly pays attention to individual aspect of speech signals but in reality, the speech signals are made up of semantics, speaker dependent features, etc. This paper therefore presents a new study that recognizes simultaneously multidimensional speaker information. In order to extract sufficient relational features, both high-level and low-level features based on series with various kinds of speech characteristics are employed in feature selection. The author builds a baseline system using support vector machine (SVM) to evaluate the correctness of multiple message identification, which contains three kinds of SVM recognizers. This system is proposed to take advantage of gender relevant messages and further the recognition performance. Experimental results show that this system using high-level feature can significantly improve the accuracy of multiple recognition by 1.46% comparing to low-level feature. And the accuracy of baseline system when using low-level feature is superior to other single classified system in some respects.