Skip to content

基于GMM与MFCC特征进行数字0-9的语音识别,GMM,MFCC,语音识别,中文数据,sklearn,Digital Voice Recognition。

Notifications You must be signed in to change notification settings

zhengyima/GMM_Digital_Voice_Recognition

Repository files navigation

GMM_Digital_Voice_Recognition

基于GMM与MFCC特征进行数字0-9的语音识别,GMM,MFCC,语音识别,中文数据,sklearn,scikit-learn,Digital Voice Recognition。

Preinstallation

 conda create -n GMM -c anaconda python=3.6 numpy pyaudio scipy #也可以使用pip
 conda activate GMM
 pip install -r requirements.txt

数据链接: https://pan.baidu.com/s/124TiAs8m7Ioa2_3dUrxGSg 提取码: xsfe

以下命令假设下载数据至/tmp/dataset.zip

Launch the script

  git clone https://github.com/zhengyima/GMM_Digital_Voice_Recognition.git GMM_DVR
  cd GMM_DVR
  unzip /tmp/dataset.zip -d ./  # dataset.zip是从百度网盘下载的数据
  python speaker-recognition.py -t enroll -i  "./data_zh_1/*/"  -m model.out
  python speaker-recognition.py -t predict  -i  "./data_zh_test_1/*/"  -m model.out

HMM实现

DTW实现

About

基于GMM与MFCC特征进行数字0-9的语音识别,GMM,MFCC,语音识别,中文数据,sklearn,Digital Voice Recognition。

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages