[TTS]Docker develop-gpu-cuda10.2-cudnn7-latest 缺失 libsndfile1-dev 和 環境參數CUDA_VISIBLE_DEVICES

[kenny-chen]

Describe the bug
Docker develop-gpu-cuda10.2-cudnn7-latest 缺失:

1. apt install libsndfile1-dev
2. 環境參數 os.environ["CUDA_VISIBLE_DEVICES"] = "0"

To Reproduce

1. 按照官方教程，使用Docker鏡像部署 docker pull paddlecloud/paddlespeech:develop-gpu-cuda10.2-cudnn7-latest
2. 進入Docker鏡像 nvidia-docker run --runtime=nvidia -it -v ~/PycharmProjects/PaddleSpeech:/home paddlecloud/paddlespeech:develop-gpu-cuda10.2-cudnn7-latest /bin/bash
3. 成功捕獲顯示卡 nvidia-smi

nvcc -V

./test_cuda
   
4. 但執行demo項目，cd demos/style_fs2 && ./run.sh 出現error

Bug

1. 缺失 libsndfile1-dev
   Traceback (most recent call last): File "style_syn.py", line 22, in <module> import soundfile as sf File "/usr/local/python3.7.0/lib/python3.7/site-packages/soundfile.py", line 192, in <module> _snd = _ffi.dlopen(_explicit_libname) OSError: cannot load library 'libsndfile.so': libsndfile.so: cannot open shared object file: No such file or directory
   我猜測，這是CUDA10.2官方docker的bug，里面缺少了libsndfile1-dev。
   當執行apt install libsndfile1-dev後，成功解決。

2. 缺失 環境參數CUDA_VISIBLE_DEVICES
   Traceback (most recent call last): File "style_syn.py", line 228, in <module> main() File "style_syn.py", line 209, in main paddle.set_device("gpu") File "/usr/local/python3.7.0/lib/python3.7/site-packages/paddle/device/__init__.py", line 313, in set_device place = _convert_to_place(device) File "/usr/local/python3.7.0/lib/python3.7/site-packages/paddle/device/__init__.py", line 204, in _convert_to_place place = core.CUDAPlace(ParallelEnv().dev_id) OSError: (External) CUDA error(100), no CUDA-capable device is detected. [Hint: 'cudaErrorNoDevice'. This indicates that no CUDA-capable devices were detected by the installed CUDA driver. ] (at /paddle/paddle/phi/backends/gpu/cuda/cuda_info.cc:66)
   因缺失 環境參數CUDA_VISIBLE_DEVICES，導致PaddleSpeech失敗獲取GPU。
   import os os.environ["CUDA_VISIBLE_DEVICES"] = "0"
   當設置CUDA_VISIBLE_DEVICES=0後，成功解決。

Environment:

• Docker: develop-gpu-cuda10.2-cudnn7-latest
• CUDA: 10.2
• CUDNN: 7

Suggestion

1. 在docker develop-gpu-cuda10.2-cudnn7-latest 鏡像 再安裝 apt install libsndfile1-dev
2. 在 /demos/style_fs2/style_syn.py 新增 import os os.environ["CUDA_VISIBLE_DEVICES"] = "0"

[lizezheng]

非常感謝提出的寶貴建議，歡迎持續關注 paddlespeech 並貢獻代碼

[kenny-chen]

非常感謝提出的寶貴建議，歡迎持續關注 paddlespeech 並貢獻代碼

你好，请问有develop-gpu-cuda10.2-cudnn7-latest的dockerfile吗?
这样我可以修改dockerfile代码。