基于内容的多声道音频检索系统开发-机械设计及理论专业论文.docxVIP

摘要摘 摘要 摘 要 现有网上音乐检索局限于类浏览和基于文字的查找功能。为了寻找一首乐曲，用户 必需准确地知道标题、作者、演奏家或者其他相关的文字描述，否则就只能花费大量时 间，逐一浏览这个音乐分类下的所有乐曲。这种基于文本的信息检索方式，脱离实际音 乐检索需要很远。例如哼起一首歌曲的主旋律却想不起歌名，很难通过传统的方法来找 到乐曲。这就需要通过对音乐本身特征的识别，实现基于内容的音乐检索。 近年来，基于内容的音频检索技术成为热点课题，对单声道音频的旋律特征提取算 法，国外不少研究机构已有卓越贡献，并开发出基于音频感知特征的音频检索商业化引 擎。我国港台地区学者改进国外已有理论和方法，取得进展。国内相关研究项目也不少， 其成果比国外和港台同类研究还有差距。 本论文工作内容是多声道音频的基于内容的旋律识别与检索。单声道的MIDI文件 可以按时间延续信息按顺序提取，多声道的MIDI音乐文件的处理则要复杂的多。在乐 理知识基础上分析多声道音频的旋律特征识别、表达方式，改进提取算法，适应多声道 处理。设计面向多声道MIDI格式文件作为音乐旋律的数据源，构建基于内容的音频检 索系统，通过人的哼唱即可检索乐曲。另外，还探索了WAV格式音乐旋律的数据源的系 统。 设计的有效性已通过原型系统实验室运行证实。 关键词： 旋律，哼唱检索，多声道，基于内容的音频检索 AbstractAbstract Abstract Abstract Currently,web music retrieval is confined in categorized scanning and text based searching．To find a song，the user has to know a series of information about its topic， composer,and performer．Otherwise，it Can be explored tediously along the category． Such a text based retrieval technology is far from satisfactory for music searching．For example，it would be extremely difficult，if one want to find a song by groaning its melody．Thus，the music retrieval should be capable of recognizing and extracting the audio features of music． Namely,the retrieval technology is to be improved from text—based to content-based． In the recent years，content-based music retrieval attracted extensive study．Fruitful results in algorithm for single·channel melody features extracting were appeared in foreign literature and relevant retrieval software was available．Researchers in Hone Kong and Taiwan districts contributed in this area by improving these theory and method．Quite a number of domestic research works were also reported．But，there is some disparity in comparison、^，im the level already achieved． Work of the theses is multi—channel music melody features’recognition and extraction． Single-channel audio information can be extracted sequentially．But,it is much more complicated for multi—channel ones．Based on music theory,representation of the melody is analyzed and extraction for mul