Skip to content

基于DTW与MFCC特征进行数字0-9的语音识别,DTW,MFCC,语音识别,中英数据,端点检测,Digital Voice Recognition。

Notifications You must be signed in to change notification settings

zhengyima/DTW_Digital_Voice_Recognition

Repository files navigation

DTW_Digital_Voice_Recognition

基于DTW与MFCC特征提取进行数字0-9的语音识别,DTW,MFCC,语音识别,端点检测,中英数据,Digital Voice Recognition。

Preinstallation

以下命令主要使用Anaconda安装环境,当然也可以更换为pip。

 conda create -n dtw -c anaconda python=3.6 numpy tqdm pyaudio scipy #也可以使用pip
 conda activate dtw
 pip install python_speech_features

Launch the script

  git clone https://github.com/zhengyima/DTW_Digital_Voice_Recognition.git DTW_DVR
  cd DTW_DVR
  mkdir ProcessedData # 创建端点处理后数据目录
  python endpointDetection_RecordedVoice.py # 端点检测,默认使用英文数据
  python VoiceRecog.py 

Links

GMM实现

HMM实现

About

基于DTW与MFCC特征进行数字0-9的语音识别,DTW,MFCC,语音识别,中英数据,端点检测,Digital Voice Recognition。

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages