基于统计学的生理期语音特征分析-《中国医学物理学杂志》

文章信息/Info

Author(s):: QIN Caimeng1; LU Tao1; LU Yu2; 1. School of Life Science, Beijing University of Chinese Medicine, Beijing 100029, China 2. Institute of Information on Traditional Chinese Medicine, China Academy of Chinese Medical Sciences, Beijing 100700, China

Keywords:: Keywords: physiological period ovulatory period menstrual period voice feature

摘要:: 目的：建立快速准确且安全无创的生理期语音监测模式。方法：采用中医闻声诊技术对育龄期女性进行生理期语音特征分析，分别以排卵期、月经期、痛经程度作为判别依据，筛选出可反映上述指征的差异性语音频率特征，拟建立相关回归模型。结果：在242项语音指标中分析得出排卵期与非排卵期、月经期与非月经期、4种痛经程度中，分别有14、11、4项语音特征指标差异具有统计学意义，其中与排卵期相关的语音特征指标为语音第五共振峰高度和最持续帧最大能量频率占时比，与月经期呈负相关的语音特征指标为最高能频段序号，与痛经程度呈正相关的语音特征指标为音阶。结论：育龄期女性在生理期语音呈现特征性变化，对排卵期、月经期具有一定的预测能力，对痛经程度的量化具有一定的参考价值。

Abstract:: Abstract: Objective To establish a rapid, accurate, safe and non-invasive mode of physiological voice monitoring. Methods The physiological voice characteristics of women in reproductive age were analyzed with auscultation and olfaction. The ovulatory period, menstrual period, and the degree of dysmenorrhea were used as the discriminatory basis to screen out the differential voice frequency characteristics that could reflect the above indications, and a regression model was established. Results Among the 242 voice indicators, there were 14, 11 and 4 voice indicators which were significantly different between ovulation and non-ovulation, menstruation and non-menstruation, and in 4 dysmenorrhea degrees, respectively. The voice resonance peak time sequence5 and the maximum energy frequency to time ratio of the most sustained frame were associated with ovulatory period and the highest energy frequency band number was negatively associated with menstrual period and the tone scale was positively correlated with the degree of dysmenorrhea. Conclusion Women of reproductive age show characteristic changes in voice during the physiological period, which has certain predictive ability for ovulation and menstruation, and has certain reference value for quantifying the degree of dysmenorrhea.