Vowel devoicing rates in Japanese from a sentence corpus
Yasuko Nagano-Madsen · 2009
The phenomenon of vowel devoicing has been described for a long time in Japanese linguistic and phonetic literature (Pohvanov 1976 [1914], Han 1962, Sawashima 1971, Hirose 1971, Sugito 1988, Maekawa 1989 etc.). In recent years there has been a growing body of data with respect to the hierarchy and interaction among the factors responsible in inducing vowel devoicing in Japanese (Takeda & Kuwabara 1987, K i m u r a et al. 1988, Yosh ida & Sagisaka 1990, Yoshida 1992, 1993, Kondo 1993, K a w a i et al. 1993). Some of these studies have been carried out for technical use such as speech synthesis and recognition and are quantitatively more adequate than previous ones. Based on the results of these recent studies, in particular on K a w a i et al . 1993, the present paper explores the extent to which the reahzation of vowel devoicing varies according to segmental contexts and accentuation using a sentence corpus. Although past studies have supplied much data on vowel devoicing factors in Japanese, potential variation due to segmental contexts and accentuation have not been explored from a sentence corpus.