摘要 |
An aspect of the present invention provides a voice recognition system includes that a memory unit (140, 270, 350,701, 801) configured to store a statistical language dictionary (706, 803) which statistically registers connections among words, a voice recognition unit (120, 220, 330, 704) configured to recognize an input voice based on the statistical language dictionary (706, 803), a prediction unit (130, 250, 330, 805) configured to predict, according to the recognition result provided by the voice recognition unit (120, 220, 330, 704), connected words possibly voiced after the input voice, and a probability changing unit (150, 260, 330, 804) configured to change the probabilities of connected words in the statistical language dictionary (706, 803) according to the prediction result provided by the prediction unit (130, 250, 330, 805), wherein the voice recognition unit (120, 220, 330, 704) recognizes next input voice based on the statistical language dictionary (706, 803) changed by the probability changing unit (150, 260, 330, 804) and wherein the memory unit (140, 270, 350, 701, 801), the voice recognition unit (120, 220, 330, 704), the prediction unit (130, 250, 330, 805) and the probability changing unit (150, 260, 330, 804) are configured to be installed in the mobile unit. <IMAGE>
|