Incomplete Transcriptions

[Nerdward]

While using the medium model on a GPU system, I get this error message:

ERROR:root:[ERROR]: Sending data to client: sent 1000 (OK); then received 1000 (OK)
This mean the client disconnected before the server could send some segments
My code:

client = TranscriptionClient(
        "localhost",
        9090,
        lang="en",
        translate=False,
        model='medium',
        use_vad=False
    )
client()

[alpcansoydas]

Same error with large-v2 model. Any solution?

[alpcansoydas]

I think we need to change the default max_connection_time=600 to higher in server.py

[makaveli10]

@alpcansoydas by incomplete you mean the transcriptions are incomplete or does it mean that it terminates after 10 minutes?

[alpcansoydas]

It terminates after 10 minutes, but it is not an error. I got it. Sorry for confusion.

[Nerdward]

@alpcansoydas by incomplete you mean the transcriptions are incomplete or does it mean that it terminates after 10 minutes?

The transcriptions are not complete during live recording. When I stop the live recording, all the transcription is not sent. I think it is a bug.