摘要
基于内容的音频描述是计算机音乐领域的一个重要的新兴研究课题。它主要被用来解决声音分离、分割、声音事件和声音源的刻划及表征、听觉场景分析和音乐分析等问题。结合音频标准讨论,介绍了基于内容的音频描述方案和描述符,以及如何对它们进行提MPEG-7取和应用。
Content-based audio description is a very important research topic in computer music industry. It aims to resolve problems such as sound segregation, segmentation, sound event and source characterization, auditory scene analysis and music analysis, etc. Combined with the ongoing MPEG-7 Audio Standard, content-based description schemes(DS) and descriptors(D) are introduced and discussed in this paper. Moreover, how to extract and apply them is also given.
出处
《计算机工程》
CAS
CSCD
北大核心
2001年第10期140-143,共4页
Computer Engineering
