Co-training Approach for Label-minimized Audio Classification

摘要：

Audio classification is an important preprocess to the audio data. However, lots of manual labeled data are needed for training models. In order to solve this problem, we evaluate a semi-supervised machine learning algorithm called co-training for content-based audio classification. The audio is divided into there classes: pure speech, pure music and speech mixed with music. We consider the audio features as views and minimize the labeled data quantity by using cotraining algorithm. The experimental results on the VOA Special English show the effectiveness of the co-training algorithm for audio classification.

关键词： Speech processing Audio classification Cotraining Label-minimaized

作者: Zhang Wei Zhao Qun Liu Yayu Pang Minhui

作者单位: Ocean University of China, Qingdao, Shandong, 266100, China

会议类型: 国际会议

会议名称: 2010 International Conference on Measuring Technology and Mechatronics Automation(ICMTMA 2010)(2010年检测技术与机电自动化国际会议)

会议地点: 长沙

会议语种:英文

页码: 860-863

在线出版日期: 2010-03-13（万方平台首次上网日期，不代表论文的发表时间）