Initial training time?

Hi all, after perfecting my Home Assistant set-up (is that ever perfect :slight_smile:) I decided to go the next step with a local voice assistant. My choice fell on Rhasspy. I followed the set-up for the Raspberry Pi (2B) with a Respeaker2 HAT last night and everything seems to be installed fine. Language is NL. I’ve started the initial training (I’ve made no modifications to the sentences file) this afternoon and after an hour the pop-up ‘Training profile’ still shows. This could be perfectly fine, however I can’t seem to find how long this should take. Should I be patient or is there something wrong?

As far as I know it should not take that long. Normally it runs for 10min max or it times out. Though the popup doesn’t matter, because you can close that at any time. Rhasspy actively doing stuff is shown by the rhasspy logo in the top left corner of the gui rotating.

Just close the popup and check the log if you see anything in there.

I also never ran rhasspy on a pi other than a pi4, so I don’t know how long the basic training takes, but an hour seems pretty long.


Might I suggest using the HA addon for Rhasspy? You can you the Pi and respeaker as a satellite

Well, that explains the pop-up at least :slight_smile: I changed the STT engine to Kaldi, redownloaded the profile and started the training again. It seems to have completed now!

However, it seems now there is some misconfiguration with the Respeaker. When I try to test it on the settings page, I receive the following error in logs:

[ERROR:2020-12-31 08:36:12,676] rhasspyserver_hermes: 
Traceback (most recent call last):
  File "/usr/lib/rhasspy/.venv/lib/python3.7/site-packages/quart/", line 1821, in full_dispatch_request
    result = await self.dispatch_request(request_context)
  File "/usr/lib/rhasspy/.venv/lib/python3.7/site-packages/quart/", line 1869, in dispatch_request
    return await handler(**request_.view_args)
  File "/usr/lib/rhasspy/rhasspy-server-hermes/rhasspyserver_hermes/", line 758, in api_test_microphones
    microphones = await core.get_microphones(test=True)
  File "/usr/lib/rhasspy/rhasspy-server-hermes/rhasspyserver_hermes/", line 825, in get_microphones
    handle_finished(), messages, message_types
  File "/usr/lib/rhasspy/rhasspy-server-hermes/rhasspyserver_hermes/", line 971, in publish_wait
    result_awaitable, timeout=timeout_seconds
  File "/usr/lib/python3.7/asyncio/", line 449, in wait_for
    raise futures.TimeoutError()

I found this in several other threads as well. Is this a recurrent problem?

I’ll look into this, thank you!