The Application Of Combining Neural Network With Information Fusion Theory In Speaker Recognition

Data Fusion, as a new front-line discipline combining traditional disciplines with newly developed engineering areas, has been extensively applied in many other areas than military ones to which it has been originally applied. This paper researches the application of combining neural network with data fusion theory Based on systematic discussion of traditional and singular feature identity recognition(such as speaker recognition and face recognition) methods, it proposes to integrate the feature parameters of face images and voice signals and use neural network, which are perception neural networks and BP network to do identity recognition. Simulative experiments are done using the fusion system and traditional ones respectively, and the result demonstrates the better function of the fusion system.The paper reviews traditional recognition method, particularly those for speaker recognition and face recognition. The face recognition process includes detection and locating and feature extraction. And, the paper introduces the theoretical background, including neural network and information fusion.In the end, on the basis of the previous information in combination, two fusion method are put forward. The first, the decision-level fusion method based on perception neural networks, it can fuse the result of speaker recognition and face recognition directly. And the second, the decision-level fusion based on BP network, the system extracts features from these two types of data, and input fusion recognition system through integrating data for recognition. By doing real-data simulative experiments, the experiment's result of fusion system is better than singular feature identity recognition.