FACEGOOD-Audio2Face

项目地址：GitHub - FACEGOOD/FACEGOOD-Audio2Face: http://www.facegood.cc

• 安装anaconda
• 安装cuda10.0
• 安装vs2017
• 安装对应版本cudnn并添加如下系统环境变量

        C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\lib\x64

        C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\include

        C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\extras\CUPTI\libx64

        C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\bin

        C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0\libnvvp

• 安装git for win  v2.36.1
• 新建环境

        conda create -n Audio2Face_env python=3.6

        python版本为3.6，然后激活环境

        conda activate Audio2Face_env

        安装tersorflow-gpu 1.15

        conda install tensorflow-gpu=1.15.0

• 安装依赖项

        pyaudio v0.2.11

        conda install -c anaconda pyaudio

        requests v2.23.0

        conda install -c conda-forge requests

        websocket v10.3

        conda install -c conda-forge websockets

        websocket-client v1.2.1

        conda install -c conda-forge websocket-client

• 按照如下步骤进行测试
• 出现如下报错

        Internal: Invoking ptxas not supported on Windows
        Relying on driver to perform ptx compilation. This message will be only logged once.

• 因此重新创建一个环境，使用cpu进行测试

        conda create -n Audio2Face_env_cpu python=3.6

• 激活环境

        conda activate Audio2Face_env_cpu

• 步骤同上，但是安装的为cpu版本

        成功运行test