Overview of Front-end Features for Robust Speaker Recognition
This paper provides an overview of automatic speaker recognition technologies, with an emphasis on front-end features for robust speaker recognition. We categorize the front-end features into low-level features and high-level features. We discussed in detail several selected low-level and high-level features including their motivations, mechanism, reported improvements, and limitations, etc. The goal of this paper is to help beginners in speaker recognition research area to catch the overall picture of current front-end features quickly.
Qin Jin Thomas Fang Zheng
Language Technologies Institute, Carnegie Mellon University, Pittsburgh, PA USA Center for Speech and Language, Tsinghua University, Beijing
国际会议
2011亚太信号与信息处理协会年度峰会(APSIPAASC 2011)
西安
英文
1-12
2011-10-18（万方平台首次上网日期，不代表论文的发表时间）