METHOD AND APPARATUS FOR CORRECTING ERROR OF RECOGNITION IN SPEECH RECOGNITION SYSTEM

摘要

A method of correcting an error in a speech recognition system of the present invention includes the steps of searching for a first correct answer candidate group for a speech recognition error using a speech recognition error-correct answer pair DB based on an acoustic model, Searching for a second correct answer candidate for a speech recognition error; searching for a third correct answer candidate for the speech recognition error using the user error correction information DB; and using the domain speech pattern DB and proper noun database Searching for a fourth correct answer candidate for the speech recognition error, and displaying the sorted correct answer candidates in each of the correct answer candidate groups.