Building Medium‐Vocabulary Isolated‐Word Lithuanian HMM Speech Recognition System

Article type: Research Article

Authors: Raškinis, Gailius | Raškinienė, Danutė

Affiliations: Faculty of Computer Science, Vytautas Magnus University, Vileikos 8, LT‐3035 Kaunas, Lithuania, e‐mail: idgara@vdu.lt | Institute of Information Technologies, Lithuanian University of Agriculture, Studentų 11, LT‐4324 Akademija, Kauno r., Lithuania, e‐mail: danra@info.lzua.lt

Abstract: In this paper, the opening work on the development of a Lithuanian HMM speech recognition system is described. The triphone single‐Gaussian HMM speech recognition system based on Mel Frequency Cepstral Coefficients (MFCC) was developed using HTK toolkit. Hidden Markov model's parameters were estimated from phone‐level hand‐annotated Lithuanian speech corpus. The system was evaluated on a speaker‐independent ∼750 distinct isolated‐word recognition task. Though the speaker adaptation and language modeling techniques were not used, the system was performing at 20% word error rate.

Keywords: speech recognition, hidden Markov modeling (HMM), HTK toolkit, hand‐annotated speech corpora

Journal: Informatica, vol. 14, no. 1, pp. 75-84, 2003