Identification of words from spoken input read from a file. See also direct speech recognition.