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 FilesNVIDIA GPU Computing ToolkitCUDAv10.0libx64
C:Program FilesNVIDIA GPU Computing ToolkitCUDAv10.0include
C:Program FilesNVIDIA GPU Computing ToolkitCUDAv10.0extrasCUPTIlibx64
C:Program FilesNVIDIA GPU Computing ToolkitCUDAv10.0bin
C:Program FilesNVIDIA GPU Computing ToolkitCUDAv10.0libnvvp
- 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
