Issue in deepspeech setup in Red Hat Linux version 2.32

#1

Hello! I am new to deepspeech and I am trying to setup deepspeech on RHEL version 2.32, after successfully doing pip install deepspeech, and taking care of virtual environments and path.

I wanted to use deepspeech python bindings, as have written a python code around deepspeech for voice to text and based it on client.py hence would want import deepspeech to work , which it doesn’t as I am getting the following error and would be grateful for a quick resolution.

Import%20Deepspeech%20error
Import Deepspeech error.jpg3892×1708 517 KB

Thank you for creating such an awesome project which has potential to be so much more.

#2

Similar issue

And if you expect help here, dont make screenshots with your mobile. It’s far easier and more convenient to just copy/paste the text here.

#3

Thank you so much for your quick reply and pointing out to the similar issue.

According to the similar issue, its the problem with CUDA dependencies, but I am not using a GPU, or even planning to train a model, I just want to run deepspeech using the pretrained model for which I am getting this error.

#4

First, please avoid sharing screenshot for debugging, it makes it very hard to read for some people.

It is not. Please carefully read the error: our libdeepspeech.so and the Python bindings are built against a more recent version glibc than yours. You will have to rebuild at least the python bindings, maybe libdeepspeech.so.

#5

Okay, first of all ,sorry for posting a screenshot and thank you for such a quick reply.

Can you please guide me how I can rebuild the python bindings and also libdeepspeech.so for a specific version of glibc that I have.

Thank you for your time.

#6

This is all documented in native_client/README.md and README.md