| 摘要 |
<P>PROBLEM TO BE SOLVED: To record reproduction start time information on divided speech data, reproduced sound source direction indication information, and keyword indexes representing the contents of the speech data additionally to individual speech files when an analog speech signal is inputted after being converted into digital speech data or inputted directly from a digital speech file and then recorded on a recording medium by dividing for each sound area after silent areas are deleted from the digital speech data. <P>SOLUTION: When deciding whether an input speech is silent or sonant, dividing it into sound areas and storing the voiced areas, the device calculates and records the start time of divided data as reproduction start time information additionally to a file, and additionally records reproduced sound source direction indication information inputted by divided data and keyword indexes of divided data extracted through speech recognition. <P>COPYRIGHT: (C)2006,JPO&NCIPI |
