VoiceStudio: a HMM-based tool for research and teaching in the speech recognition field [articol]

Abstract

This paper introduces the Romanian speech recognition system VoiceStudio. As most state-of-the-art Automatic Speech Recognition (ASR) systems today, it is based on Hidden Markov Models. Although there are numerous toolkits designed for this task, they usually have no visual interface, which means that the student or the researcher needs to spend some considerably amount of time in order to learn their functionality. The system’s modular design, together with some implementation issues are pointed out, as well as the future plans of development.