Handler exception after no audio i assume

[florian-asche]

Hi,
i am using wyoming-satellite without vad. I think after i speak the wake word, but nothing else there is a timeout problem inside the wyoming-satellite. That is documented here: rhasspy/wyoming-satellite#237.

But what's not documented is that also wyoming-faster-whisper is crashing and no longer working until i manually restart it.

I am using the latest docker based version.

Error Log:

INFO:__main__:Ready
INFO:faster_whisper:Processing audio with duration 00:03.300
INFO:wyoming_faster_whisper.handler: Wie spät ist es?
INFO:faster_whisper:Processing audio with duration 00:03.980
INFO:wyoming_faster_whisper.handler: Kannst du meine Geschichte vorlesen?
INFO:faster_whisper:Processing audio with duration 00:04.750
INFO:wyoming_faster_whisper.handler: Ich versuche zu verstehen, was du für ein Problem hast.
ERROR:asyncio:Task exception was never retrieved
future: <Task finished name='wyoming event handler' coro=<AsyncEventHandler.run() done, defined at /usr/local/lib/python3.9/dist-packages/wyoming/server.py:31> exception=AssertionError()>
Traceback (most recent call last):
  File "/usr/local/lib/python3.9/dist-packages/wyoming/server.py", line 41, in run
    if not (await self.handle_event(event)):
  File "/usr/local/lib/python3.9/dist-packages/wyoming_faster_whisper/handler.py", line 63, in handle_event
    assert self._wav_file is not None
AssertionError

[TheRealPSV]

Did you ever find a workaround for this? I tried to find a way to implement a healthcheck, but it seems like wyoming is actually still running somehow, so if I run something like echo '{ "type": "describe" }' | nc -w 1 localhost 10300 inside the container, I still get a valid wyoming response. I can't find a way to actually test within the container if whisper has crashed.

[synesthesiam]

Is this still a problem with the Linux Voice Assistant? https://github.com/OHF-Voice/linux-voice-assistant

[TheRealPSV]

Unfortunately, my system is set up to do all voice processing off-device for speed, so I can't really try out the linux voice assistant.