×

Multilingual speech recognition

  • US 7,716,050 B2
  • Filed: 11/17/2003
  • Issued: 05/11/2010
  • Est. Priority Date: 11/15/2002
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented method in which a computer system initiates execution of software instructions stored in memory, the computer-implemented method comprising:

  • accepting text spellings of training words in a plurality of sets of training words, each set corresponding to a different one of a plurality of languages;

    for each of the sets of training words in the plurality, receiving pronunciations for the training words in the set, the pronunciations being characteristic of native speakers of the language of the set, the pronunciations also being in terms of subword units at least some of which are common to two or more of the languages; and

    training a single pronunciation estimator using data comprising the text spellings and the pronunciations of the training words; and

    calculating a single acoustic subword model for each subword unit, based on the pronunciations in the plurality of sets of training words, by mixing distributions of acoustic parameters representing the sounds of the subword unit in multiple languages when a subword unit is common to two or more languages.

View all claims
  • 8 Assignments
Timeline View
Assignment View
    ×
    ×