文档介绍：基于余弦距离的哼唱音乐检索算法的设计与实现
学院
计算机学院
专业
计算机科学与技术
班级
学号
姓名
指导教师
负责教师
2010年6月
摘要
随着多媒体技术的发展,网络音乐日益增多。现在人们己经不满足于通过歌曲名、歌曲的演唱者等一些文本信息来检索。特别是对于那些种类繁多的音乐数据,人们也许只记得一个调子,也许只记得一个片断,如何快速有效的通过旋律来检索相关音乐就成为一个突出问题。本文对基于哼唱的音乐检索进行了相关的研究和实现。
本文采用了Microsoft Visual C++编程技术和MySQL server数据库,做了如下工作:
通过对音乐信号基本理论的研究,提出了利用音高差和音长来描述音乐旋律特征的方法,有效地避免哼唱过程走调问题,因此提高了准确性。
对于哼唱音乐片段,通过信号预处理、基音提取、特征提取、特征后处理,实现了从WAV文件中提取音符音高差和音长特征,采用n_gram方法对音高序列进行切分,扩充查询集合以提高查询准确度。
确定WAV格式作为数据库音乐的文件存储格式,并建立歌曲的音高差和音长特征库。
为提高查询速度,建立整数编码的哈希索引。
采用余弦距离旋律匹配算法,从音高差和音长两方面进行旋律匹配,返回查询结果,并计算响应时间。
关键词:哼唱检索;特征提取;余弦距离匹配;音高差USD和音长旋律特征
Design and implement of hummed melody retrieval based on consine distance
Abstract
With rapid development of multimedia technology, a large number of online music can be obtained from the . Now, people are no longer satisfied with searching music based on text information such as music name and singer. Especially, with various kinds of music data, people might only remember the tune of a song, or perhaps fragment of it. How to quickly and effectively retrieve music data by melody information has e a prominent problem. In this dissertation, the music retrieval based on hummed melody is studied .
In this dissertation, the hummed melody retrieval based on consine distance is implemented with Microsoft Visual C++ and MySQL server, and the concrete work is as follows:
Firstly, through the study of music signal, proposing that pitch interval and duracion as the music melody feature, which is effective to avoid the pitch changing as well as improving the music retrieval accuracy. Secondly, for the hummed melody, extracting the feature of music melody from WAV music’s format through signal preprocessing, keynote extraction, feature extraction, feature extraction postprocessing. Using n_gram to syncopate the pitch interval in order to improve the accuracy. Thirdly, making sure that WAV as the music’s format and set up pitch interval and duracion database. Forthly, in order to improve the speed