An Automatic Approach Towards Audio Segmentation and Classification

摘要：

In this paper, we present an automatic audio segmentation and classification system. First, a new feature based on ICA is proposed to improve the multiple change-point segmentation method. The false alarm rate has significant improvements over original algorithm applied MFCCs. In audio classification, this study adopts three features from different domains to classify each clip into its belonged types based on threshold and majority rule is applied to determine the final type. The superiority of the proposed system has been demonstrated via a 400-file audio database with a high accuracy rate.

关键词： Audio segmentation and classification ICA Feature extraction Threshold

作者: Wenjuan Pan Zongwu Wang Zhijing Liu

作者单位: the Research Center of Computer Information Applications, Xidian University, Xi’an 710071, China Xidian University, Xi’an 710071, China

会议类型: 国际会议

会议名称: The Second International Symposium on Intelligence Computation and Applications(ISICA 2007)(第二届智能计算及其应用国际会议)

会议地点: 武汉

会议语种:英文

在线出版日期: 2007-09-21（万方平台首次上网日期，不代表论文的发表时间）