Estimating high‐confidence portions based on agreement among outputs of multiple LVCSR models

Abstract

AbstractThis paper experimentally evaluates the agreement among the outputs of multiple Japanese LVCSR models, with respect to whether it is effective as an estimate of confidence for each hypothesized word. The results of experimental evaluation show that the agreement between the outputs with two LVCSR models with different decoders and acoustic models can achieve quite reliable confidence. Furthermore, among various features of acoustic models based on Gaussian mixture HMMs, it is concluded that ones such as whether or not to have short pause models, as well as different units in HMMs are the most effective in achieving highly reliable confidence. © 2004 Wiley Periodicals, Inc. Syst Comp Jpn, 35(7): 33–40, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.10636