ERIC Number: ED218943
Record Type: Non-Journal
Publication Date: 1981
Reference Count: N/A
Speech Synthesis Applied to Language Teaching.
Studies in Language Learning, v3 n1 p171-81 Spr 1981
The experimental addition of speech output to computer-based Esperanto lessons using speech synthesized from text is described. Because of Esperanto's phonetic spelling and simple rhythm, it is particularly easy to describe the mechanisms of Esperanto synthesis. Attention is directed to how the text-to-speech conversion is performed and the ways in which this kind of speech output has been exploited to provide novel kinds of learning situations. The apparatus outputs various sounds according to numerical commands. It not only produces the sounds but also produces the necessary transitions from one sound to the next. Esperanto is written phonetically, with each letter corresponding to just one sound, independent of environment (allophonic variation is relatively slight in the spoken language). Word stress is always on the next-to-last vowel, and monosyllables are almost always unstressed function words. These factors greatly simplify the task of converting text to speech. The conversion of Esperanto text to speech is performed with the aid of a table which contains for each input letter not only the corresponding synthesizer phoneme code(s) but also numbers which specify special subroutines to be invoked for certain letters. The PLATO system is one of the few computer systems which makes explicit provision for diacritics. A critical advantage of speech synthesis in computer-based education is the ability to generate a message as needed by the program. In the case of dictation drills, beginning students have difficulty, mainly due to failings of the synthesizer in producing good consonants. The quality of the synthetic speech is likely to be improved in the next few years. (SW)
Descriptors: Artificial Languages, Artificial Speech, Auditory Perception, College Second Language Programs, Computer Assisted Instruction, Higher Education, Online Systems, Phonetics, Phonology, Programed Instructional Materials, Second Language Instruction, Speech, Teaching Methods
Not available separately, see FL 012 990.
Publication Type: Reports - Descriptive; Guides - Non-Classroom; Journal Articles
Education Level: N/A
Authoring Institution: Illinois Univ., Urbana. Language Learning Lab.