Continuous speech recognition without end‐point detection

Abstract

AbstractA new continuous speech recognition method that does not need explicit speech end‐point detection is proposed. A one‐pass decoding algorithm is modified to decode the input speech of infinite length so that, with appropriate nonspeech models for silence and ambient noises, continuous speech recognition can be executed without explicit end‐point detection. The basic algorithm (1) decodes a processing block of the predetermined length, (2) backtraces and finds the boundaries of the processing blocks where the word history in the preceding processing block is merged into one, and (3) restarts decoding from the boundary frame with the merged word history. The effectiveness of the method is verified by spoken dialogue transcription experiments. With a 5‐minute dialogue in a moving car, the proposed method gives better results in word accuracy than the results using the explicit end‐point detection method and the conventional one‐pass decoder. © 2006 Wiley Periodicals, Inc. Electr Eng Jpn, 156(4): 43–50, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/eej.20140