摘要
基于音频的场景分析技术对机器自动感知环境特征,提高自动化程度等具有重要的意义。考虑到人耳的听觉特性,采用反映人类听觉特性的Mel倒谱系数MFCC(Melfrequencycep-stralcoefficient)作为音频信号特征,提出了一种用矢量量化来进行音频场景分析的方法。
The techniques of audio scene analysis are very important for the automatic awareness of background events and improving the automation.In this paper,the auditory charac-ters of human being are considering,thus,the MFCC(Mel frequency cepstral coefficient),which represents the auditory characters,is adopted as the feature of audio signal,and a method of au-dio scene analysis based on vector quantization is proposed.
出处
《电声技术》
北大核心
2002年第3期8-10,共3页
Audio Engineering
基金
哈尔滨工业大学跨学科交叉研究基金(HIT.MD.200001)资助
关键词
矢量量化
音频信号
场景分析
语音信号处理
audio signal
scene analysis
vector quantization
Mel frequency cepstral coef-ficient
