Automatic speech recognition (ASR) has reached very high levels of performance in controlled situations. However, the performance degrades significantly when environmental noise ...
Abstract. We examine the problem of message recognition by reviewing the definitions and the security model in the literature. In particular, we examine the Jane Doe protocol, whi...
Recognizing graphic objects from binary images is an important task in many real-life applications. Generally, there are two ways to do the graphics recognition: onestep methods a...
In the Pattern Recognition field, growing interest has been shown in recent years for Multiple Classifier Systems and particularly for Bagging, Boosting and Random Subspaces. Th...
Discriminative training has been a leading factor for improving automatic speech recognition (ASR) performance over the last decade. The traditional discriminative training, howev...