server.py Errors when running on docker

[nachiketh89]

Hello Respected Scholars,
I am trying to reproduce the steps you have mentioned in the GitHub - Open-Speech-EkStep/speech-recognition-open-api wesbite :-

I am running the following command :-
docker run -itd -p 50051:50051 --env gpu=False --env languages=['en','hi'] -v C:\Users\nachi\Downloads\EKSTEP\speech-recognition-open-api\deployed_models:/opt/speech_recognition_open_api/deployed_models/ gcr.io/ekstepspeechrecognition/speech_recognition_model_api:3.2.37

In the Windows 11 docker log , I see the following :-

2023-09-13 23:50:58 /usr/local/lib/python3.8/dist-packages/torch/cuda/init.py:52: UserWarning: CUDA initialization: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx (Triggered internally at /pytorch/c10/cuda/CUDAFunctions.cpp:100.)
2023-09-13 23:50:58 return torch._C._cuda_getDeviceCount() > 0
2023-09-13 23:50:58 [NeMo W 2023-09-13 18:20:58 optimizers:46] Apex was not found. Using the lamb optimizer will error out.
2023-09-13 23:50:59 No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda'
2023-09-13 23:51:01 2023-09-13 18:21:01,732 — [MainThread] - src.lib.inference_lib - inference_lib.py.get_cuda_device(56) - INFO - User has provided gpu as False gpu_present False
2023-09-13 23:51:01 Using server workers: 10
2023-09-13 23:51:01 2023-09-13 18:21:01,780 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(29) - INFO - Initializing realtime and batch inference service
2023-09-13 23:51:01 2023-09-13 18:21:01,781 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(38) - INFO - User has provided gpu as False
2023-09-13 23:51:01 2023-09-13 18:21:01,781 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(41) - INFO - GPU available on machine False
2023-09-13 23:51:01 2023-09-13 18:21:01,782 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(44) - INFO - Loading models from /opt/speech_recognition_open_api/deployed_models/ with gpu value: False
2023-09-13 23:51:01 2023-09-13 18:21:01,782 — [MainThread] - src.model_service - model_service.py.init(35) - INFO - environment requested languages ['en','hi']
2023-09-13 23:51:01 Traceback (most recent call last):
2023-09-13 23:51:01 File "/opt/speech_recognition_open_api/server.py", line 24, in
2023-09-13 23:51:01 run()
2023-09-13 23:51:01 File "/opt/speech_recognition_open_api/server.py", line 17, in run
2023-09-13 23:51:01 add_SpeechRecognizerServicer_to_server(SpeechRecognizer(), server)
2023-09-13 23:51:01 File "/opt/speech_recognition_open_api/src/speech_recognition_service.py", line 45, in init
2023-09-13 23:51:01 self.model_service = ModelService(self.MODEL_BASE_PATH, 'kenlm', gpu, gpu)
2023-09-13 23:51:01 File "/opt/speech_recognition_open_api/src/model_service.py", line 39, in init
2023-09-13 23:51:01 model_config = json.load(f)
2023-09-13 23:51:01 File "/usr/lib/python3.8/json/init.py", line 293, in load
2023-09-13 23:51:01 return loads(fp.read(),
2023-09-13 23:51:01 File "/usr/lib/python3.8/json/init.py", line 357, in loads
2023-09-13 23:51:01 return _default_decoder.decode(s)
2023-09-13 23:51:01 File "/usr/lib/python3.8/json/decoder.py", line 337, in decode
2023-09-13 23:51:01 obj, end = self.raw_decode(s, idx=_w(s, 0).end())
2023-09-13 23:51:01 File "/usr/lib/python3.8/json/decoder.py", line 353, in raw_decode
2023-09-13 23:51:01 obj, end = self.scan_once(s, idx)
2023-09-13 23:51:01 json.decoder.JSONDecodeError: Invalid \escape: line 3 column 20 (char 33)

When I tried with GPUs TRUE , I get the below errors :-

I have submitted " docker run -itd -p 50051:50051 --env gpu=True --env languages=['en','hi'] --gpus all -v C:\Users\nachi\Downloads\EKSTEP\speech-recognition-open-api\deployed_models:/opt/speech_recognition_open_api/deployed_models/ gcr.io/ekstepspeechrecognition/speech_recognition_model_api:3.2.37 " on Win11 cmd in administrator mode , and I see the below errors :-

=========
2023-09-14 00:11:20 [NeMo W 2023-09-13 18:41:20 optimizers:46] Apex was not found. Using the lamb optimizer will error out.
2023-09-14 00:11:23 2023-09-13 18:41:23,658 — [MainThread] - src.lib.inference_lib - inference_lib.py.get_cuda_device(56) - INFO - User has provided gpu as True gpu_present True
2023-09-14 00:11:23 2023-09-13 18:41:23,659 — [MainThread] - src.lib.inference_lib - inference_lib.py.get_cuda_device(61) - INFO - ### GPU Utilization ###
2023-09-14 00:11:23 | ID | GPU | MEM |
2023-09-14 00:11:23 ------------------
2023-09-14 00:11:23 | 0 | 0% | 17% |
2023-09-14 00:11:23 2023-09-13 18:41:23,839 — [MainThread] - src.lib.inference_lib - inference_lib.py.get_cuda_device(67) - INFO - available GPUs ['0'], all GPUs [0], excluded GPUs [0]
2023-09-14 00:11:23 2023-09-13 18:41:23,906 — [MainThread] - src.lib.inference_lib - inference_lib.py.get_cuda_device(75) - INFO - Selected GPUs: [] requested GPUs [0]
2023-09-14 00:11:23 2023-09-13 18:41:23,907 — [MainThread] - src.lib.inference_lib - inference_lib.py.get_cuda_device(82) - INFO - selected gpu index: None selecting device: cuda
2023-09-14 00:11:23 Using server workers: 10
2023-09-14 00:11:23 2023-09-13 18:41:23,966 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(29) - INFO - Initializing realtime and batch inference service
2023-09-14 00:11:23 2023-09-13 18:41:23,966 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(38) - INFO - User has provided gpu as True
2023-09-14 00:11:23 2023-09-13 18:41:23,967 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(41) - INFO - GPU available on machine True
2023-09-14 00:11:23 2023-09-13 18:41:23,968 — [MainThread] - src.speech_recognition_service - speech_recognition_service.py.init(44) - INFO - Loading models from /opt/speech_recognition_open_api/deployed_models/ with gpu value: True
2023-09-14 00:11:23 2023-09-13 18:41:23,968 — [MainThread] - src.model_service - model_service.py.init(35) - INFO - environment requested languages ['en','hi']
2023-09-14 00:11:23 Traceback (most recent call last):
2023-09-14 00:11:23 File "/opt/speech_recognition_open_api/server.py", line 24, in
2023-09-14 00:11:23 run()
2023-09-14 00:11:23 File "/opt/speech_recognition_open_api/server.py", line 17, in run
2023-09-14 00:11:23 add_SpeechRecognizerServicer_to_server(SpeechRecognizer(), server)
2023-09-14 00:11:23 File "/opt/speech_recognition_open_api/src/speech_recognition_service.py", line 45, in init
2023-09-14 00:11:23 self.model_service = ModelService(self.MODEL_BASE_PATH, 'kenlm', gpu, gpu)
2023-09-14 00:11:23 File "/opt/speech_recognition_open_api/src/model_service.py", line 39, in init
2023-09-14 00:11:23 model_config = json.load(f)
2023-09-14 00:11:23 File "/usr/lib/python3.8/json/init.py", line 293, in load
2023-09-14 00:11:23 return loads(fp.read(),
2023-09-14 00:11:23 File "/usr/lib/python3.8/json/init.py", line 357, in loads
2023-09-14 00:11:23 return _default_decoder.decode(s)
2023-09-14 00:11:23 File "/usr/lib/python3.8/json/decoder.py", line 337, in decode
2023-09-14 00:11:23 obj, end = self.raw_decode(s, idx=_w(s, 0).end())
2023-09-14 00:11:23 File "/usr/lib/python3.8/json/decoder.py", line 353, in raw_decode
2023-09-14 00:11:23 obj, end = self.scan_once(s, idx)
2023-09-14 00:11:23 json.decoder.JSONDecodeError: Invalid \escape: line 3 column 20 (char 33)

Any clue to the issue with server.py will help me greatly in getting your system up and running .

I beg you to please help me sir .

My WhatsApp number is +91-8861636108 , if you can share you WhatsApp number or mobile numbers, it would greatly help to get around this issue.
My email id is [email protected] ,,,,, Pls help me ,, I have got stuck !

Thanking You ,
Nachiketh