日本传统“诗吟”中单个反射声优选时延与自相关函数的关系(英文)

Preferred Delay Time of the Single Reflection for Japanese Traditional"Shigin"Singing in Relation to the Autocorrelation Function

研究在日本传统"诗吟"表演中,在干信号中加上适当时延的单个反射声是否更为有利.专业"诗吟"表演者嗓音中对典型的"节"的吟唱是该种表演艺术的一种独特的吟咏方式.对此及对"节"外内容的吟唱都在消音室中加以录音.文中通过成对比较法试验得出了主观优选的标度值.结果表明,在聆听对"节"的吟唱时,听者更喜欢在干信号中加上延时为10～30ms的单个反射声,而在聆听对"节"外内容的吟唱时,加上单个反射声与否对听者而言并无显著的差别.文中对此项发现从吟唱"节"的声源信号的自相关函数的包络特征上加以讨论.

This study investigates whether the added single reflection with its delay is preferred for Japanese traditional ＂ Shigin （詩吟） ＂ singing. As the dry sources, an expert ＂Shigin＂ performer＇s vowel singings tor a typical phrase with ＂Fushi （ 節 ） ＂—a special style of intonation proper to ＂Shigin＂ performance—and those without ＂Fushi＂ were recorded in an anechoic chamber. The paired-comparison test was then conducted to obtain the scale values of subjective preference. The results show that listeners prefer the added single reflection with its delay in the range of 10 to 30 ms for the singing with ＂Fushi＂, and that no clear difference of preference among the added single reflection with its delay exists for the singing without ＂ Fushi＂. The findings are discussed in relation to the characteristic envelope of the running autocorrelation function of the singing voice with ＂Fushi＂.