Automatic speech recognition system for Turkish spoken language

The transmission and storage of speech sounds is possible for decades. In addition by usingsignal processing techniques, it is also possible to process speech signals. By using timeand frequency analysis of speech signal and several machine learning algorithms, it ispossible to build a system which is used to recognize spoken words. Such systems arecalled Automatic Speech Recognition systems.In our work, we have used the Automatic Speech Recognition system for Turkish spokenlanguage which has built by BUSIM speech group. However, the output of the recognizer isthe list of spoken words. Even for humans it is a very hard task to understand a text withoutpunctuation symbols. Hence to build more complex recognizer whose goal to perform topicsegmentation and topic summarization, the output of ASR should be divided into sentencesat first.Our goal is to build a system which performs the sentence segmentation. In our work wehave used ASR system to obtain word level and phoneme level time marks and by usingthat time marks with the audio files, we have extracted prosodic features, where theprosodic properties of speech contains information about the punctuation in the text, whichis not available at the output of ASR system.