搜索资源列表
mfccdtw
- 先用端点检测将语音中有用的语音部分提取出来(即将头部和尾部的静音部分除掉),然后用LPC算法提取语音信号的特征参数,进行动态归整(DTW算法)后与模板库里面的标准语音作比较,最后将识别结果进行D/A转化后播放出来。在本部分的设计中,则主要完成语音识别的模式匹配算法部分的软件实现。 -First with the endpoint detection of speech to voice some of the useful extracted from the (soon to mute som
voice-recognition_matlab-code
- 读入语音文件,并对其做时域、频域的分析,提取相关特征参数。进行线性预测分析,得到LPC谱等线性预测参数,并做了基于预测误差的基音周期估计。-read .wav files,analysing them in time domain,frequency domain and extract some feature parameters related,then do linear prediction analysis ,and get LPC linear prediction paramet
LPC
- 利用LPC算法实现对于语音信号的特征提取,将一帧的语音信号用简化模型的参数来表示-The use of the LPC algorithm to extract to achieve For the the characteristics of of the voice signal, will the the a of the voice to represent the by the signal with a simplified the the the parameters of of