ASRT_SpeechRecognition 本地环境搭建

请记住标红的永远是最重要的，就够了，这两个版本不对类似也白搭， python版本是3.6

 

conda env list

conda create -n voice

conda  activate voice

conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free

conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/

conda config --set show_channel_urls yes

python decoder.py

pip install soundfile
   

pip install python_speech_features

pip install tqdm

pip install easydict
   

python

import tensorflow as tf

tf.__version__

nvcc --version

tensorflow DLL load failed: 找不到指定的模块
pip uninstall tensorflow   也就是卸载tensorflow。

（2）卸载完成之后再重新安装，输入
pip install tensorflow 

conda  activate voice
pip install tensorflow==1.13.1   -i https://mirrors.aliyun.com/pypi/simple/
pip install tensorflow-gpu==1.13.1   -i https://mirrors.aliyun.com/pypi/simple/

pip uninstall tensorflow   也就是卸载tensorflow。

pip install tensorflow==2.2.0   -i https://mirrors.aliyun.com/pypi/simple/

pip uninstall tensorflow -y
pip uninstall keras -y

pip show tensorflow

pip search tensorflow

pip install tensorflow==1.15.0   -i https://mirrors.aliyun.com/pypi/simple/

pip install keras==2.2.4

 
 1.18.1

1023  wget https://www.python.org/ftp/python/3.6.5/Python-3.6.5.tgz
 1024  tar -zxvf Python-3.6.5.tgz
 1025  cd Python-3.6.5
 1026  ./configure
 1027  make
 1028  make install
 1029  python3 -V
 1030  mv /usr/bin/python /usr/bin/python.bak
 1031  ln -s /usr/local/bin/python3 /usr/bin/python
 1032  vi /usr/bin/yum
 1033  cd /usr/bin/
 1034  python:
 1035  vi /usr/libexec/urlgrabber-ext-down
 1036  ll
 1037  pwd

1052  /usr/local/bin
 1053  ln -s /usr/local/bin/pip3 /usr/bin/pip
 1054  rm -rf /usr/bin/pip
 1055  ln -s /usr/local/bin/pip3 /usr/bin/pip
 1056  pip
 1058  
 1059  
 1060  
 1061  history
tar zxf data_thchs30.tgz
tar zxf ST-CMDS-20170001_1-OS.tar.gz
D:\work\dstc\python\ASRT_SpeechRecognition\model_speech\m251\speech_model251_e_0_step_500.model

pip install --upgrade pip
pip install tensorflow==2.0.0a0   -i https://mirrors.aliyun.com/pypi/simple/

pip install keras==2.3.1   -i https://mirrors.aliyun.com/pypi/simple/

pip install numpy
pip install matplotlib
pip install scipy
pip install wave 
 pip install math 
 pip install h5py 

 pip install python_speech_features

 pip install http 
 pip install urllib 
 pip uninstall  numpy
pip install numpy==1.18.1

conda install tensorflow=2.1 -y
conda install keras=2.3.1 -y
conda install matplotlib -y
conda install python_speech_features -y
conda install matplotlib -y
conda install matplotlib -y

NotImplementedError: eval is not supported when eager execution is enabled, is .numpy() what you're looking f

我tensorflow_gpu版本是2.1.0；报如下错误：
NotImplementedError: eval is not supported when eager execution is enabled, is .numpy() what you’re looking for?
在引入tensorflow包后，加上这句：

tf.compat.v1.disable_eager_execution()