|
Phonetic-acoustic database of Russian trigrams
Yu. I. Butenko, Yu. V. Stroganov, A. V. Kvasnikov, N. V. Slavnov N. E. Bauman Moscow State Technical University, 5-1, 2nd Baumanskaya Str., Moscow 105005, Russian Federation
Abstract:
The article describes the phonetic-acoustic base of Russian trigrams for analysis and synthesis of Russian speech. The classification of the Russian trigrams is given as well as trigrams easy and difficult for pronunciation are highlighted. It is noted that the trigrams in the composition of the word fully or partially coincide with the morphemes of the Russian language. The variants of marking of speech records in the system of marking sounding speech are illustrated. Variability in pronunciation of Russian trigrams by different speakers is analyzed and illustrated by means of oscillograms. It is shown that the speech markup system allows taking into account personal characteristics of the speaker, affecting the quality of pronunciation. The influence of phoneme location in the word on the quality of its recognition is studied. It is suggested to use frequency of use and the position of the phoneme in the word as weights when using trigrams in speech recognition and synthesis tasks.
Keywords:
phonetic-acoustic base, trigram, speaker, annotation, oscillogram, pronunciation, variability.
Received: 18.08.2020
Citation:
Yu. I. Butenko, Yu. V. Stroganov, A. V. Kvasnikov, N. V. Slavnov, “Phonetic-acoustic database of Russian trigrams”, Sistemy i Sredstva Inform., 32:1 (2022), 55–62
Linking options:
https://www.mathnet.ru/eng/ssi811 https://www.mathnet.ru/eng/ssi/v32/i1/p55
|
Statistics & downloads: |
Abstract page: | 76 | Full-text PDF : | 34 | References: | 18 |
|