Text-to-speech synthesis

• Main Authors: Gapuz, Josephine, Lim, Derwin O., Ong, Aileen L., Wu, Hsin Hsin
• Format: text
• Language: English
• Published: Animo Repository 1993
• Subjects: Speech synthesis; Computer input-output equipment; Coding theory; Programming (Electronic computers); Text processing (Computer science)
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/16381
• Institution: De La Salle University

Text-to-Speech synthesis is primarily concerned with the conversion of text into a corresponding speech output. The whole system is made up of hardware and software modules which run hand-in-hand to produce the required audio output. The hardware module is made up of a speech synthesizer circuitry which provides the necessary pre-recorded allophone codes. The software module is responsible for the actual conversion of text into allophonic codes, which is the basis of the hardware-software interface. Allophone stringing was performed by a special set of rules patterned after the algorithms used in NRL, LINGUA, and MITALK. Syllabication, intonation, duration analysis, and pause insertions were also used to improve the quality of the speech output. Although the direct phoneme synthesis method of artificial speech synthesis should provide an unlimited vocabulary, a truly unlimited speech capability is unattainable. The conversion routine is guided by a set of rules based on the English language, which is generally considered to be complex. Text-to-Speech synthesis has proved to be of use to many applications. In fact, the future may find it a useful companion to speech recognition systems, or expand the system to speak in other languages, and include sound effects and music.