| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To detect a changing position of a speaker in a speech data created by recording a speech sentence etc. for language learning. <P>SOLUTION: A speech data processing system 1 comprises: a speech data acquiring section 12 for acquiring the speech data created by recording speech of a plurality of persons one by one; a classification section 13 for applying predetermined pattern recognition processing on the speech data, and classifying the speech data into a part for each person who utters the voice, based on the pattern recognition processing result; and a position determination section 14 for determining the changing position of the speaker in the speech data, based on the classification result by the classification section 13. <P>COPYRIGHT: (C)2009,JPO&INPIT</p> |
