一个基于语谱图映射的音乐检索系统
首发时间:2011-12-29
摘要:随着计算机的普及和互联网的发展,数字音乐得到广泛的传播。在浩如烟海的音乐库中如何快速、有效的寻找到所需要的音乐是一项非常有意义的课题。目前商业搜索引擎如谷歌、百度等依靠音乐相关的文字信息完成检索任务,由于其固有的缺陷难以满足用户的新需求。本文采用改进的语谱图映射算法,对查询样例的语谱图进行矢量映射并提取特征点,利用哈希结构建立索引,采用多级检索方法,返回最相似音频。通过实验,该系统的准确率和速度都比较让人满意。
For information in English, please click here
A Music Information Retrieval System Based On Projections Of Spectrograms
Abstract:With the popularization of computers and development of the Internet, digital music spread widely. How to find out the needed music quickly and efficiently in the massive library, is a significant subject. So far, commercial search engines such as Google and Baidu and so on, implement retrieval through character information related to music. Because of its intrinsic limitation, this method is inadequate to meet the new demand of users.Algorithm of projections of spectrograms is improved, spectrograms of sample are mapped to vectors and feature points are extracted. Indices are established using hash structure, the most appropriate audio is returned using multilevel retrieval. Through experiments, the accuracy and efficiency of this system are satisfactory.
Keywords: music retrieval music fingerprinting projections of spectrograms
基金:
论文图表:
引用
No.****
同行评议
共计0人参与
勘误表
一个基于语谱图映射的音乐检索系统
评论
全部评论0/1000