期刊
2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
卷 -, 期 -, 页码 4618-4622出版社
IEEE
DOI: 10.1109/ICASSP43922.2022.9746378
关键词
Speaker age estimation; Label distribution learning; Variance regularization; Attribute inference
资金
- Key Research and Development Program of Guangdong Province [2021B0101400003]
- National Key Research and Development Program of China [2018YFB0204403]
Existing methods for speaker age estimation face challenges caused by label ambiguity. This paper proposes a method that utilizes label distribution learning to combine age classification and regression approaches, achieving improved performance on a real-world dataset.
Existing methods for speaker age estimation usually treat it as a multi-class classification or a regression problem. However, precise age identification remains a challenge due to label ambiguity, i.e., utterances from adjacent age of the same person are often indistinguishable. To address this, we utilize the ambiguous information among the age labels, convert each age label into a discrete label distribution and leverage the label distribution learning (LDL) method to fit the data. For each audio data sample, our method produces a age distribution of its speaker, and on top of the distribution we also perform two other tasks: age prediction and age uncertainty minimization. Therefore, our method naturally combines the age classification and regression approaches, which enhances the robustness of our method. We conduct experiments on the public NIST SRE08-10 dataset and a real-world dataset, which exhibit that our method outperforms baseline methods by a relatively large margin, yielding a 10% reduction in terms of mean absolute error (MAE) on a real-world dataset.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据