form one of the largest language group and these languages are very close in morphological and syntactic levels. These languages have rich morphological structures (it is possible to form very large number of word-forms from a given stem by adding the rich set of simple and compound suffixes) and as a result it is very difficult to create an MT system for… (More)
This article describes the main stages of the development and characteristics of a speech recognition system for the Azerbaijani language developed within the project Dilmanc. Some numerical data on the amount of information to create the database for normal functioning of the developed software modules and some characteristics of a speech recognition… (More)
Any algorithm for speech recognition always has a procedure for comparison of an input signal with the etalons available in memory, for definition of its belongings to an appropriate class. In this paper the definition of the set of etalons for the 1st version of the Automatic Speech Recognition (ASR) system for the Azerbaijani language is described.