Research on Matching Method in Humming Retrieval

摘要：

　　With the development of science and technology,computer technology is increasingly updated,and multimedia search technology has been widely used.For music retrieval,text-based retrieval technology cant meet the diversified retrieval needs.Humming retrieval is through matching the input audio and the audio in the database to match the audio.Its a more convenient way to retrieve music.In this paper the dynamic time wrapping matching method,hidden markov model and location sensitive hash algorithm in humming retrieval were systematically introduced,and the location sensitive hash algorithm with the dynamic time wrapping algorithm was combined to improve the retrieval rate effectively.

关键词： Humming Retrieval DTW HMM LSH

作者: Shaojing Zhou Zhijun Zhao Ping Shi Mingliang Han

作者单位: College of Information Engineering,Communication University of China

会议类型: 国际会议

会议名称: 2017 IEEE 3rd Information Technology and Mechatronics Engineering Conference(ITOEC2017)(2017 IEEE 第3届信息技术与机电一体化工程国际学术会议)

会议地点: 重庆

会议语种:英文

页码: 516-520

在线出版日期: 2017-10-03（万方平台首次上网日期，不代表论文的发表时间）