The Lower Limits, Minimum Number of Frames, for Segment Selection
The lower limit, 5, for t is defined here according to our experiment results. Take the sentence, /na-2 bu-4 shu-1 ren-4 hə-2 zie-3 zyue-2 fang-1 an-4/, as an example. When the lower limit for t is set to 1, 3, and 5, the speech segments automatically selected will be as those shown in Fig. A, Fig. B, and Fig. C, respectively. From Fig. A, it can be seen that several selected segments are of very short durations. Also, there is still a difference in the selected segments between Fig. B and Fig. C.

Fig. A    

Fig. B    

Fig. C