PASLĖPTŲJŲ MARKOVO MODELIŲ METODO TYRIMAS IR TAIKYMAS BALSO ĮRAŠAMS STENOGRAFUOTI
INVESTIGATION OF HIDDEN MARKOV MODELS AND ADAPTATION FOR TRANSCRIBING VOICE RECORDINGS
Author(s): Jūratė Vaičiulytė, Gražvydas FelinskasSubject(s): Phonetics / Phonology, Baltic Languages, ICT Information and Communications Technologies
Published by: Vilniaus Universiteto Leidykla
Keywords: speech recognition; language models; acoustic models; hidden Markov models;
Summary/Abstract: Automatic speech recognition methods have been reviewed in the paper. Hidden Markov models for acoustic modeling were used, a language model and dictionary were created. The models were trained with recorded news, total 5859 words, for 2 hours and 20 minutes. Two phonetic systems (with 85 and 32 phonemes respectively) were developed on the basis of a Lithuanian pronunciation dictionary. The created models were integrated in a stenography app prototype for transcribing voice recordings. The testing results of the acoustic models showed a 94% accuracy for phrase recognition.
Journal: Jaunųjų mokslininkų darbai
- Issue Year: 2016
- Issue No: 1 (45)
- Page Range: 71-78
- Page Count: 8
- Language: Lithuanian