Text Conditioning and Statistical Language Modeling Aspects for Romanian Language

Abstract: In this paper we present a synthesis of the theoretical fundamentals and some practical aspects of statistical (n-gram) language modeling which is a main part of a large vocabulary statistical speech recognition system. There are presented the unigram, bigram and trigram language models as well as the add one, Witten-Bell and Good-Turing estimator based Katz back-off smoothing algorithms. The perplexity measure of a language model used for evaluation is also described. The practical experiments were made on Romanian Constitution corpus. Text normalization steps before the language model generation are also presented. The results are ARPA-MIT format language models for Romanian language. The models were tested and compared using perplexity measure. Finally some conclusions are drawn based on the experimental results