Title :
Significance of durational knowledge for speech synthesis system in an Indian language
Author :
Kumar, S. R Rajesh ; Yegnanarayana, B.
Author_Institution :
Dept. of Comput. Sci. & Eng., Indian Inst. of Technol., Madras, India
Abstract :
A discussion is presented of the issues involved in the design and development of a speech synthesis system for the Indian language, Hindi. A speech synthesis system is being developed based on the parameter concatenation model and having the flexibility to incorporate prosodic features. The characters of Hindi have been chosen as the basic units. A framework for collecting the basic units has been evolved, keeping in mind the flexibility needed to incorporate prosodic rules. The importance of incorporating durational knowledge for speech synthesis in Hindi is demonstrated. In order to improve the quality of speech produced from input text, it is necessary to expand the many variations of durations of segments that occur in natural continuous speech in Hindi
Keywords :
speech synthesis; Hindi; Indian language; durational knowledge; linear prediction model; natural continuous speech; parameter concatenation model; prosodic features; speech synthesis; text-to-speech synthesis; Acoustical engineering; Computer science; Data mining; Natural languages; Predictive models; Rhythm; Spatial databases; Speech coding; Speech synthesis; Stress;
Conference_Titel :
TENCON '89. Fourth IEEE Region 10 International Conference
Conference_Location :
Bombay
DOI :
10.1109/TENCON.1989.176983
