×

SPEECH RECOGNITION SYSTEM AND PROGRAM THEREFOR

  • US 20100057457A1
  • Filed: 11/30/2007
  • Published: 03/04/2010
  • Est. Priority Date: 11/30/2006
  • Status: Active Grant
First Claim
Patent Images

1. A speech recognition system comprising:

  • a speech recognition section that converts speech data into text data by using a speech recognition dictionary containing a large volume of word pronunciation data each constituted by a combination of a word and one or more corresponding pronunciations, each pronunciation including one or more phonemes, and that has a function of adding to the text data a start time and a finish time of a word segment in the speech data corresponding to each word included in text data;

    a word correcting section that presents competitive candidates for each word in the text data acquired from the speech recognition section, allows each word to be corrected by selecting a correct word from among the competitive candidates for correction if the correct word is included in the competitive candidates, or by manually inputting a correct word if no correct word is included in the competitive candidates;

    a phoneme sequence converting section that recognizes the speech data in units of phoneme, converts the recognized speech data into a phoneme sequence composed of a plurality of phonemes, and that has a function of adding to the phoneme sequence a start time and a finish time of each phoneme unit in the speech data corresponding to each phoneme included in the phoneme sequence;

    a phoneme sequence extracting section that extracts from the phoneme sequence a phoneme sequence portion composed of one or more phonemes existing in a segment corresponding to a period of the start time and finish time of the word segment of a word corrected by the word correcting section;

    a pronunciation determining section that determines the phoneme sequence portion as the pronunciation of the word corrected by the word correcting section; and

    an additional registration section that combines the corrected word with the pronunciation determined by the pronunciation determining section as new word pronunciation data and additionally registers the new word pronunciation data in the speech recognition dictionary if it is determined that the corrected word has not been registered in the speech recognition dictionary, or additionally registers the pronunciation determined by the pronunciation determining section in the speech recognition dictionary as another pronunciation of the corrected word if it is determined that the corrected word is a registered word that has already been registered in the speech recognition dictionary.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×