The ´neural´ phonetic typewriter

Author

Kohonen, Teuvo

Author_Institution

Lab. of Comput. & Inf. Sci., Helsinki Univ. of Technol., Espoo, Finland

Volume

21

Issue

3

fYear

1988

fDate

3/1/1988 12:00:00 AM

Firstpage

11

Lastpage

22

Abstract

The factors that make speech recognition difficult are examined, and the potential of neural computers for this purpose is discussed. A speaker-adaptive system that transcribes dictation using an unlimited vocabulary is presented that is based on a neural network processor for the recognition of phonetic units of speech. The acoustic preprocessing, vector quantization, neural network model, and shortcut learning algorithm used are described. The utilization of phonotopic maps and of postprocessing in symbolic forms are discussed. Hardware implementations and performance of the neural networks are considered.<>

Keywords

artificial intelligence; neural nets; speech analysis and processing; speech recognition; acoustic preprocessing; artificial intelligence; neural computers; neural network processor; phonotopic maps; shortcut learning algorithm; speaker-adaptive system; speech recognition; symbolic forms; vector quantization; Automatic speech recognition; Natural languages; Neural networks; Optical films; Pattern recognition; Speech recognition; Speech synthesis; Telephony; Text recognition; Vocabulary;

fLanguage

English

Journal_Title

Computer

Publisher

ieee

ISSN

0018-9162

Type

jour

DOI

10.1109/2.28

Filename

28