|
Sasang constitution experts utilize voice characteristics as an auxiliary measure for deciding a person's constitutional group. This study aims at establishing relationship between speech features and the constitutional groups by subjective listening evaluation of voice characteristics. A speech database of 841 speakers whose constitutional groups have been already diagnosed by Sasang constitution experts is constructed. Speech features related to speech source and vocal tract filter are extracted from five vowels and a sentence. Statistically significant speech features for classifying the groups are analyzed using the SPSS. The features contributed to constitution classification are speaking rate, Energy, A1, A2, A3, H1, H2, H4, CPP for 20s male, F0_mean, CPP, SPI, HNR, Shimmer, Energy, A1, A2, A3, H1, H2, H4 for 20s female, Energy, A1, A2, A3, H1, H2, H4, CPP for 60s male, and Jitter, HNR, CPP, SPI for 60s female. Experimental results show that speech technology are useful in classifying the constitutional groups. |
