Computer Models for Speech Understanding
Research on Automatic Speech Recognition (ASR) began in the early fifties by the attempt to extract significant features from acoustic data and to classify them, using methodologies developed in the area of Pattern Recognition.
KeywordsEurope Transportation Editing Extractor Hyde
Unable to display preview. Download preview PDF.
- Fundamental books dealing with speech problems are those by Fant (1960, 1973) and Flanagan (1972). Google Scholar
- An excellent review of the work done in automatic speech recognition before the seventies has been done by Hyde (1972) in an interesting book by David and Denes (1972). Google Scholar
- Design concepts for Speech Understanding Systems are given in the book by Newell et al. (1973). A good collection of tutorial papers for Speech Understanding can be found in the book by Reddy (1975). Recent results of important projects are described in books by Lea (1980), Walker (1978), Dixon and Martin (1979) and in reports by Lea and Soup (1979), Woods et al. (1976), Reddy (1976). Google Scholar
- Interesting works for relating Phonetics, Phonology, and Perception models with Automatic Speech Recognition are due to O’Malley (1976), White (1976), Oshika et al. (1975), Cohen and Mercer (1975); fundamental in this area are also the books by Chomsky and Halle (1968) and Fant (1973).Google Scholar
- Recent reviews of the state of the art in the field have been provided by Reddy (1976), Martin (1976), Jelinek (1976), Klatt (1977), Medress et al. (1977), Zagoruiko (1977), De Mori (1979), Erman et al. (1980). Google Scholar
- An excellent review of the work done in Japan is contained in the Ph.D. thesis by Nakagawa (1976), available in English. Interesting monographs describe the work done in the Soviet-Union; among them are worth mentioning those by Tsiemiel (1971), Trunin-Donskoi (1975), Fain (1977), Zagoruiko (1976), Chitavichius (1977). Google Scholar
- Progress in France is reviewed in the thesis by P. Quinton (1980).Google Scholar
- A recent paper by Kaplan (1980) describes available industrial products and the most promising laboratory prototypes and research items.Google Scholar
- Of particular interest are also the progress reports of the Massachussetts Institute of Technology (RLE) in Cambridge, Mass., the Haskins Laboratories in New Haven, Conn., the Electrotechnical Laboratory in Tokyo, the Royal Institute of Technology in Stockholm.Google Scholar
- There are many journals in which the research results in Speech Understanding can be found. Among them, the IEEE Transactions on Acoustic, Speech and Signal Processing (ASSP), on Pattern Analysis and Machine Intelligence (PAMI), the Journal of Problems for Information Transmission, Signal Processing, Pattern Recognition, Artificial Intelligence, the International Journal on Man-Machine Studies and the forthcoming journal entitled Speech Communication.Google Scholar
- Among the International Conferences, those on Pattern Recognition and those on Artificial Intelligence usually contain papers on Speech Understanding. Other important conferences are ICASSP (USA), ARSO (USSR), EUSIPCO (Europe), CALF (France).Google Scholar