Method and apparatus for improving the transcription accuracy of speech recognition software

  • US 7,805,299 B2
  • Filed: 10/23/2006
  • Issued: 09/28/2010
  • Est. Priority Date: 03/01/2004
  • Status: Active Grant
  • ×
    • Pin Icon | RPX Insight
    • Pin
First Claim
Patent Images

1. A method for operating a computerized, speech recognition system, said speech recognition system including a default vocabulary database and an adjunct vocabulary database, said adjunct vocabulary database having vocabulary elements previously used within said speech recognition system, said vocabulary elements within said adjunct vocabulary database including at least one tag associated with a previous use of said vocabulary element within said system, said tag having associated weighting information, said method comprising:

  • providing speech input to said speech recognition system, said speech recognition system providing a set of vocabulary matching elements based on said speech input, said set of vocabulary matching elements having associated default weightings;

    creating a virtual sub-vocabulary from said adjunct vocabulary database based on said tag;

    modifying said default weightings of said set of vocabulary matching elements according to said weighting information associated with said tags; and

    selecting a vocabulary element from said set of vocabulary matching elements as a match for said speech input based on said modified weighting within said modified, set of vocabulary matching elements.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×