Talker age estimation using machine learning.
Talker age estimation using machine learning.
复制标题
DOI：
10.1121/2.0000921
复制
发表时间：
2017-06-01
期刊：
影响因子：
--
通讯作者：
Ferguson, Sarah H
中科院分区：
文献类型：
--
作者：
Berardi, Mark L;Hunter, Eric J;Ferguson, Sarah H
随着年龄的增长，声音的声学特征也会发生变化。了解一个人的声音是如何随着年龄的变化而变化的，这可能有助于我们深入了解与发声功能相关的生理变化。以前的工作已经显示了声学参数随时间年龄的变化，以及听者感知年龄和时间年龄之间的差异。然而，之前的大部分工作都是使用横截面语音样本完成的，这些样本将显示出随着年龄的变化，但可能会平均掉与个体年龄差异有关的重要变异性。目前的研究使用的是纵向录音样本，收集自一个人的演讲语料库，时间跨度约为50岁(48岁至97岁)。这项研究使用时序年龄和感知年龄作为自变量，调查了声音如何随年龄变化；感知年龄数据是在先前的直接年龄估计研究中获得的。利用纵向记录，提取了一系列语音和语音声学参数。这些参数被适用于监督学习模型，以预测按时间顺序排列的年龄和感知年龄。将讨论年龄段和感知年龄段模型之间的差异以及各种声学参数的有用性。
As a person ages, the acoustic characteristics of the voice change. Understanding how the sound of a voice changes with age may give insight into physiological changes related to vocal function. Previous work has shown changes in acoustical parameters with chronological age, as well as differences between listener-perceived age and chronological age. However, much of this previous work was done using cross-sectional speech samples, which will show changes with age but may average out important variability with regard to individual aging differences. The current study used a longitudinal recording sample gathered from a corpus of speeches from a single individual spanning about 50 years (48 to 97 years of age). This study investigates how the voice changes with age using both chronological age and perceived age as independent variables; perceived age data were obtained in a previous direct age estimation study. Using the longitudinal recordings, a range of voice and speech acoustic parameters were extracted. These parameters were fitted to a supervised learning model to predict chronological age and perceived age. Differences between the chronological age and perceived age models as well as the usefulness of the various acoustic parameters will be discussed.