Investigation of coarticulation in continuous speech of Japanese
スポンサーリンク
概要
- 論文の詳細を見る
This study analyzed coarticulation involved in continuous speech based on articulo-graphic data obtained from three Japanese male speakers. Distribution of the articulation points of vowels and consonants revealed that speakers might compensate for morphological differences in the hard palate by adjusting the location of vowel articulation points. To evaluate the effects of the left and right phonemes on the target phoneme, three-phoneme sequences, consisting of five Japanese vowels (V) and ten apical and two palatal consonants (C), were extracted from read sentences and used in this analysis. A stepwise multiple regression method was used to analyze the phoneme sequences, in order to evaluate the "contributions" of the surrounding phonemes to the central one. The results showed that the horizontal component of the articulatory movement had a dominant function during articulation. The movement of the tongue tip was highly correlated to the tongue dorsum movement in the horizontal dimension, but was almost independent in the vertical dimension. This phenomenon suggested that coarticulation in VCV sequences can be considered as an independent consonantal gesture superimposed on a transitional portion between vowels. For apical-vowel combinations, the preceding consonant in CVC had a stronger effect on the vowel than the following one, but there was no dominance caused by the positions of the vowels in VCV sequences. For palatal-vowel combinations, the following phoneme showed a greater effect than the preceding phoneme in CVC sequences. In VCV sequences, the open and closed vowels showed different behaviors.
- 社団法人日本音響学会の論文
著者
-
Dang Jianwu
Japan Advanced Inst. Of Sci. And Technol. Ishikawa Jpn
-
Dang J
Japan Advanced Institute Of Science And Technology
-
HONDA Kiyoshi
ATR Human Information Science Laboratories
-
Honda K
Atr Human Information Sci. Lab. Kyoto Jpn
-
Honda Masaaki
School of Sport Science, Waseda University
-
Honda Masaaki
School Of Sport Science Waseda University
-
Honda Kiyoshi
Atr Human Information Processing Research Laboratories
-
Honda Masaki
School of Sport Science, Waseda University
関連論文
- Robust voice activity detection based on noise eigenspace
- A bone-conduction system for auditory stimulation in MRI
- Exploring Human Speech Production Mechanisms by MRI(Speech Dynamics by Ear, Eye, Mouth and Machine)
- A model-based investigation of activations of the tongue muscles in vowel production
- Difference in vocal tract shape between upright and supine postures : Observations by an open-type MRI scanner
- Speech Enhancement based on Noise Eigenspace Projection
- A speech enhancement framework based on noise eigenspace projection (音声)
- MRI同期サンプル法による音声生成の研究
- Robust speech feature extraction based on auditory neuronal adaptation mechanism
- A Model-Based Learning Process for Modeling Coarticulation of Human Speech(Knowledge, Information and Creativity Support System)
- Normalization of vocal tract shape using radial basis function (音声)
- Optimization and Evaluation of a Coarticulation Model based on Observation and Simulation
- Parameter Optimization for a Coarticulation Model Based on Observation and Simulation (国際ワークショップ Frontiers in Speech and Hearing Research)
- Extraction of Low Dimensional Representation of Vowels in Articulatory Space (国際ワークショップ Frontiers in Speech and Hearing Research)
- Comparison of Emotion Perception among Different Cultures
- A physiological articulatory model for simulating speech production process
- Investigation of coarticulation in continuous speech of Japanese
- Investigation of coarticulation effects on vocal tract shapes of vowels based on similarity
- An experimental study of acoustic characteristics of hypopharyngeal cavities using vocal tract solid models
- Evolution of vowel production studies and observation techniques( THE VOWEL : From Chiba and Kajiyama to current issues)
- Individual variation of the hypopharyngeal cavities and its acoustic effects
- A method of tooth superimposition on MRI data for accurate measurement of vocal tract shape and dimensions