| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To improve a speech recognition rate while suppressing cost and processing time required for speech recognition processing. <P>SOLUTION: A speech recognition device 1 includes a speech acquisition unit 2, a storing unit 3, a speech recognition unit 4, a determination unit 5 and a creation unit 6. The speech recognition unit 4 acquires a first recognition rate by recognizing a speech data acquired by the speech acquisition unit 2. The determination unit 5 determines that a recognition result of the speech data is to be output, when the first recognition rate is equal to or more than a predetermined threshold. The creation unit 6 creates a correction speech signal data in which the speech data in the storing unit 3 is corrected, based on an analysis result of the speech data in the storing unit 3, when the first recognition rate is smaller than the predetermined threshold. The speech recognition unit 4 acquires a second recognition rate by recognizing the correction speech data. Then, the determination unit 5 compares the second recognition rate with the first recognition rate, to determine that the recognition result of the speech data is output, when the recognition rate at the higher side is acquired. <P>COPYRIGHT: (C)2011,JPO&INPIT</p> |
