AUTOMATIC STRESSING OF LITHUANIAN TEXT USING DECISION TREES

Abstract

This paper deals with one of the speech synthesizer components – automatic stressing of a text. The method, which by means of a decision tree finds sequences of letters that unambiguously define the word stressing, was applied to stress a Lithuanian text. Stressing rules based on sequences of letters at the beginnings, endings and in the middle of a word have been formulated. Also, the proposed method was compared with the method based on the morphological analysis, and it was proved that both methods gave similar results. The algorithm proposed in the paper reaches the accuracy of about 95.5%.