| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To reduce the burden of a user who edits a time series of feature quantities of voice signals. <P>SOLUTION: An extraction value acquisition part 42 acquires a time series of extraction values A of feature quantities relating to voice prosody. A relative value control part 54 sets a time series of relative values C, based on the extraction values A. An editing value calculating part 56 calculates a time series of editing values V, resulting from varying the extraction values A, according to the relative values C. A display control part 46 displays a relative value image GC representing the time series of relative values, on a display device 24. A control point setting part 52 sets the position of a control point PC, in a direction of a numerical axis of the relative values C out of the relative value image GC, variably in accordance with a user's indication. A relative value control part 54 sets respective relative values C at a plurality of points of time near the control point PC, according to the position of the control point PC in the direction of the number axis. <P>COPYRIGHT: (C)2010,JPO&INPIT</p> |
