Title :
Analysis of context-dependent segmental duration for automatic speech recognition
Author :
Wang, Xue ; Pols, Louis C W ; Ten Bosch, Louis F M
Author_Institution :
Inst. of Phonetic Sci., Amsterdam Univ., Netherlands
Abstract :
The paper presents statistical analyses of context dependent phone durations using the hand segmented TIMIT database, for the purpose of improving automatic speech recognition. Two main approaches were used. (1) Duration distributions were found under the influence of individual contextual factors, such as broader classes specified by long or short vowels, word stress, syllable position within the word and within an utterance, post vocalic consonants, and utterance speaking rate. (2) A hierarchically structured analysis of variance was used to study the numerical contributions of 11 different contextual factors to the variation in duration. Several systematic effects were found, whereas several others were obscured by the inherent variability in this speech material. We suggest implementation of this knowledge in the post processing phase of a recogniser
Keywords :
speech processing; speech recognition; statistical analysis; word processing; automatic speech recognition; context dependent phone durations; context dependent segmental duration; contextual factors; duration distributions; hand segmented TIMIT database; hierarchically structured analysis of variance; post processing phase; post vocalic consonants; speech material; statistical analyses; syllable position; systematic effects; utterance speaking rate; vowels; word stress; Analysis of variance; Automatic speech recognition; Character generation; Data analysis; Databases; Hidden Markov models; Speech analysis; Speech recognition; Statistical analysis; Stress;
Conference_Titel :
Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on
Conference_Location :
Philadelphia, PA
Print_ISBN :
0-7803-3555-4
DOI :
10.1109/ICSLP.1996.607818