摘要
语音处理与语音辩识是当前计算机及控制领域的热门课题。数字化语音文件的出现为该领域的研究提供了新的途径和新的方便。本文通过对数字语音文件的时域、频域分析及数据处理实验,比较其不同的音效结果,观察其频率组合成分的变化,从中得出一些重要结论,并为说话人识别等语音处理的进一步研究提供依据。
It is a significant subject in computer and control field to analyze,process and distinguish the speech. As a result of data speech files, it brings a new way and convenience to research, in this paper, we find some important conclusions by analyzing and processing the data in speech file, comparing different results and looking into the change of spectrum ,it is possible to offer some bases for more processing about the speech data, such as distinguishing who is speaking and so on.
出处
《北京电子科技学院学报》
2005年第4期61-63,75,共4页
Journal of Beijing Electronic Science And Technology Institute
关键词
语音数据
分析
处理
辨识
speech data
analysis
processes
distinguish
