F0 Feature Analysis of Communication Between Elderly Individuals for Health Assessment

F0 Feature Analysis of Communication Between Elderly Individuals for Health Assessment
复制标题

老年人健康评估沟通的F0特征分析

DOI:
10.1007/978-3-319-58466-9_19
复制
发表时间:
2017
期刊:
Human Computer Interaction @ Springer International Publishing 2017
影响因子:
--
通讯作者:
Shunpei Matsumoto
Shunpei Matsumoto
中科院分区:
--
文献类型:
--
作者:
Yumi Wakita;Shunpei Matsumoto

文献摘要

相似文献

本研究探讨了一种利用日常对话中的非语言信息来估计老年人健康状况的系统。我们已经证实了使用基频(F0)来估计年轻人之间对话气氛的有效性。与不平滑的会话相比,平滑的会话具有F0的平均值(Ave-F0)略微增加的趋势,并且其标准偏差值(SD-F0)显著增加。当使用t检验时,差异显著,其中置信水平为95%。我们证实了Ave-F0和SD-F0是有用的分离笑声话语从平常的speechutters.In本文中,我们报告的声学分析结果之间的自由交谈的老年人,并比较它与年轻人的分析结果。我们描述了使用F0特征估计健康状况的可能性。
This study explores a system that estimates the health condition of an elderly individual using nonverbal information from daily conversations.We have already confirmed the effectiveness of using the fundamental frequency (F0) to estimate the atmosphere of a conversation between young individuals. A smooth conversation has a tendency for its average value of F0 (Ave-F0) to increase slightly, and its standard deviation value (SD-F0) to increase significantly compared with a non-smooth conversation. The differences are significant when using a t-test, where the confidence level is 95%. We confirmed that Ave-F0 and SD-F0 are useful in separating laughter utterances from usual speech utterances.In this paper, we report on the acoustic analysis results of a free conversation between elderly individuals, and compare it with the analysis results of young individuals. We describe the possibility of estimating a health condition using F0 characteristics.