The normal convention with Dockerfiles is to put them in the root directory of the main project. Perhaps this should be done here?

With your docker container running, if I then do:

curl -X POST --data-binary @foo.wav http://localhost:8080/stt

... where foo.wav is a wave file with some speech, I get:

curl: (56) Recv failure: Connection reset by peer

... and I see nothing in the output on the deepspeech-server container (beyond the initial startup messages).

If I run the exact same commands in the dockerfile, just in a virtualenv, natively on my machine (Debian bullseye/testing), and then start up the server, then it works fine; the server responds with the expected text from the wave file.

Any ideas? I'm a bit new to docker, so I'm not sure what exactly about the environment is different.

hi, if i tried to use deepspeech-gpu instead, it throws the error as below

ImportError: libcublas.so.8.0: cannot open shared object file: No such file or directory

I am using nvidia-docker

I've been trying to get this working but I receive a strange error on startup:
Top line(s):
deepspeech-server | Exception in callback None()
deepspeech-server | handle: <Handle cancelled>
deepspeech-server | Traceback (most recent call last):
deepspeech-server | File "/usr/local/lib/python3.7/asyncio/events.py", line 88, in _run

Final line:
deepspeech-server | AttributeError: 'x' object has no attribute 'lm_alpha'

I noticed that it's been a while since the last update and I can see that the latest python is being pulled. Is there a possible issue with versions?