Feature extraction and speech labeling for continuous speech recognition [articol]

Abstract

Speech analysis and speech labeling are two important issues of any speech recognition system. Due to the fact that speech is analyzed on short - time intervals it is hard to make a visual interface for analysis. On the other hand it is known that speech recognition gives better results if signal’s features are used for labeling. The program described in this paper is an attempt to realize an interface for speech analysis and a feasible framework for speech labeling. The application is developed in Visual C++ 6.0, and allows the extraction and visualization of speech features frame by frame and speech labeling at word and phone level.