This paper describes a method of selecting phonetically balanced word sets and some features of voiceless plosives in various phonetic environments obtained from the phonetically balanced word set. There are complicated variations of acoustic characteristics of speech in different phonetic environments. To study those variations, word sets are needed which efficiently contain various phonetic environments. This paper describes a method of selecting word sets which are phonetically balanced and contain as many different phonetic environments as possible, in as few words as possible. From Shin-Meikai Japanese-language dictionary, a word set of 492 words was selected which had almost all VCVs in the dictionary. Some examples of acoustic characteristics variations in the acoustic-phonetic segment representation are shown. That is, formant movements in voiceless plosives of the word set were examined in various phonetic environment and their distributions in various contexts are shown.
|ジャーナル||Denshi Gijutsu Sogo Kenkyusho Iho/Bulletin of the Electrotechnical Laboratory|
|出版ステータス||Published - 1988|
ASJC Scopus subject areas