Method for recognizing utterances of a non native speaker in a speech processing system

摘要

A speech/voice recognition device (1) has a phoneme-based voice/speech model (2a) for recognition of spoken comments, remarks etc in a first language and a second phoneme-based speech/voice model (2b) for recognition of spoken remarks, comments etc in a second language. A selection device (3) automatically selects the first or second voice/speech model (2a,2b) on the basis of spoken remarks and comments of the user, where that voice/speech model (2a,2b) having a better recognition result of spoken phonemes of this user is selected. A transmission device (4) transmits phoneme sequences spoken with the characteristics of the second language and recognised by the second speech/voice model (2b) to words in the first language, on the basis of the first speech/voice model (2a), when the second voice/speech model (2b) has been selected for voice recognition, so that words in the first language are recognised when spoken with the accent of the second language.