Bangla Broadcast Speech Recognition Using Support Vector Machine

摘要

Over the past few decades, incredible growth has been revealed in the recognition of modern language at speech technologies and in its related fields alike signal processing, pattern recognition, machine learning, cognitive science, and so on. Yet the significant working progress of Bangla speech processing is not visible enough. This paper introduces the identification of Bangla broadcast speech by the use of support vector machine (SVM). This method is based on linear kernel SVM applied on MATLAB platform. Four types of noisy broadcast speech data are used in this experiment and SVM shows a fairly good outcome.