Current status and prospects of speech processing technology in Korea

We introduce the speech processing technology being studied and developed in Korea as well as its current status, problems, and future prospects. Time-varying and nonlinear methods will play important roles in the future, and considering that speech analysis is the basis of speech signal processing, this area will be studied more intensively in the future. In speech synthesis, we have already achieved comparatively good speech intelligibility, but there is room for the improvement of speech naturalness. Continuous speech recognizers which are not affected by the environment or the speakers will be constructed using HMM and ANN. For speech coding, speech coders with lower bit rate will be developed. In this area, many researchers in Korea are concentrating on the reduction of the computational burden and processing delay