In [1] the same group of authors worked on the system for discrete digits recognition in
Serbian language. Problems in that work are located, and many of them overcome and …

Speech signal processing and modeling in systems for continuous speech recognition and
Text-to-Speech synthesis in Serbian language are described in this paper. Both systems are …

This paper shortly presents program package for continuous speech recognition which is, so
far, successful with small and medium dictionaries. Package is very large because it …

U radu je prikazan sistem za prepoznavanje povezano izgovorenih reči zasnovan na
skrivenim Markovljevim modelima (eng. Hidden Markov Models-HMM). Osobenosti ovoga …

The paper presents the results of the research into the possibility of application of decision
trees to lexical stress localisation in speech corpora. The importance of solving this problem …

Automatic Speech Recognition (ASR) has a goal to train computers to understand human
speech. On the other side, Text-To-Speech (TTS) synthesis has to learn computers to read …

Automatic speech recognition and text-to-speech conversion are language dependent
technologies, developed for a small number of languages. ASR and TTS for Serbian have …

This paper shortly presents a program package for continuous speech recognition, named
AlfaNumCASR. This HMM-based phoneme in context recognizer is successful with small …