| 摘要 |
A car navigation system is provided with a voice recognition capability to search position information by means of voice recognition without requiring registration of voice recognition standard patterns for words to be recognized. <??>The system includes: a sound analysis means (113) for acoustically analyzing a user's vocal utterance and outputting an input-sound feature vector, an acoustic-model storage means (114) for storing in advance respective acoustic models for predetermined sound units in which a syllable or a phoneme is deemed a sound unit, a sound-unit recognition means (115) for checking the input-sound feature vector against the acoustic models to output a relevant sound-unit recognition candidate string, a word-and-position-information registration means (118) for correlating and registering in a word-and-position-information correlation dictionary the sound-unit recognition candidate string and position information acquired from the car navigation system main unit, and a position-information searching/outputting means (119) for calculating the acoustic likelihood by checking the input-sound feature vector outputted by the sound analysis means against sound feature vectors for each sound-unit recognition candidate string in the word-and-position-information correlation dictionary, and outputting, to the car navigation main unit, position information associated with the sound-unit recognition candidate string for which the calculated acoustic likelihood is not less than a predetermined threshold. <IMAGE> |
