嗓音研究在语言学、语音学、言语康复、嗓音评价、言语工程以及声纹鉴定等领域中都是重要的研究课题。为了进行嗓音研究我们建立了一个大规模的嗓音数据库,并对库中的嗓音样本进行嗓音参数提取和分析,在此基础上建立起的嗓音参数常模将为嗓音评价奠定数据基础。本研究以变声期的嗓音为切入点进行分析,分析涵盖了变声前期、中期、中后期以及完全稳定期四个年龄段的嗓音数据,分别为:少年阶段(11岁到15岁)、青少年阶段(16岁到18岁)、青年阶段(19到30岁)和青壮年阶段(3I岁到45岁)。本研究对每个发音样本提取了基频(F。)、绝对频率抖动(Jita)、频率抖动百分比(Jitt)、振幅抖动(ShdB)和振幅抖动百分比(Shim)共五种参数。通过大数据统计分析得出了一系列随着年龄增长的声学参数变化曲线和分布图,该数据对病理嗓音的诊断以及在声纹鉴定中,说话者年龄段的鉴别都具有积极的意义。
Voice study plays an important role in the fields of linguistics, phonetics, speech rehabilitation and verification of voiceprint. We have constructed a big database of sustained-vowels of different ages, and analyzed the data of four periods of adolescent, that is, at the beginning, in the middle, in the end and after voice change period of adolescent. Parameters include fundamental frequency, Jita, Jitt, ShdB and Shim. Finally, the relationship among ages, gender, fundamental frequency, jitter and shimmer has been found. The results would be helpful in voice assessment, verification of voiceprint, and speech engineering.