[TTS] cuda问题

[ChengsongLu]

$ paddlespeech tts --input "你好，欢迎使用百度飞桨深度学习框架！" --output output.wav [nltk_data] Error loading averaged_perceptron_tagger: <urlopen error [nltk_data] [Errno 111] Connection refused> [nltk_data] Error loading cmudict: <urlopen error [Errno 111] [nltk_data] Connection refused> /home/luchengsong/anaconda3/envs/paddle/lib/python3.9/site-packages/_distutils_hack/init.py:33: UserWarning: Setuptools is replacing distutils. warnings.warn("Setuptools is replacing distutils.") W1010 09:50:56.247031 1756 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.6, Driver API Version: 12.2, Runtime API Version: 11.8 W1010 09:50:56.247234 1756 dynamic_loader.cc:303] The third-party dynamic library (libcudnn.so) that Paddle depends on is not configured correctly. (error code is /usr/local/cuda/lib64/libcudnn.so: cannot open shared object file: No such file or directory) Suggestions:

1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
2. Configure third-party dynamic library environment variables as follows:
   • Linux: set LD_LIBRARY_PATH by export LD_LIBRARY_PATH=...
   • Windows: set PATH by `set PATH=XXX; RuntimeError: (PreconditionNotMet) Cannot load cudnn shared library. Cannot invoke method cudnnGetVersion. [Hint: cudnn_dso_handle should not be null.] (at ../paddle/phi/backends/dynload/cudnn.cc:64)

CUDA版本：12.2 paddlepaddle-gpu: 2.5.1.post120

[syg1996419]

I have the same problem

[qingjiaozyn]

我也遇到了这个问题，请问解决了吗

[lxowalle]

我在运行paddlespeech_server时也遇到了问题，原因是系统在本地找不到动态库libcudnn.so，需要自行添加库路径，我的解决方法是：

1. 查找系统中libcudnn.so的路径:sudo find / -name "libcudnn*"
2. 我搜索到路径:/home/owen/miniconda3/pkgs/cudnn-7.6.5-cuda10.2_0/lib/libcudnn.so存在libcudann.so
3. 添加库搜索路径:export LD_LIBRARY_PATH=/home/owen/miniconda3/envs/paddlespeech/lib/
4. 再次运行，不会报错了

[ZHUHF123]

paddlepaddle2.5.1 不支持cuda12.2版本 我的也这样 需要降低cuda版本 好像不支持12.*的 要11.6和11.8