Generating speech model involves successively reducing body of text on text data in user-specific second body of text, generating values of speech model using reduced first body of text

摘要

The method involves successively reducing a first body of text (10) by one or more parts depending on text data in a user-specific second body of text (11) and generating the values of a speech model (7) using the reduced first body of text (12). Unigram frequencies in the first body of text, the reduced first body of text and the second body of text are evaluated to determine the parts by which the first body of text is reduced. Independent claims are also included for the following: a speech recognition system.