2023-11-02 19:51:39.604 ERROR (MainThread) [homeassistant.components.assist_pipeline.pipeline] Unexpected error during wake-word-detection
Traceback (most recent call last):
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/assist_pipeline/pipeline.py", line 661, in wake_word_detection
result = await self.wake_word_entity.async_process_audio_stream(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/wake_word/__init__.py", line 112, in async_process_audio_stream
result = await self._async_process_audio_stream(stream, wake_word_id)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/wyoming/wake_word.py", line 152, in _async_process_audio_stream
chunk_info = audio_task.result()
^^^^^^^^^^^^^^^^^^^
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/wyoming/wake_word.py", line 82, in next_chunk
async for chunk_bytes in stream:
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/assist_pipeline/pipeline.py", line 745, in _wake_word_audio_stream
async for chunk in audio_stream:
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/assist_pipeline/pipeline.py", line 1156, in process_enhance_audio
async for dirty_samples in audio_stream:
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/esphome/voice_assistant.py", line 155, in _iterate_packets
raise RuntimeError("Not running")
RuntimeError: Not running
#I keep running into this error when trying to get an M5 atom echo working
delicate wadi
regal heath
Not same hardware but same error
https://github.com/home-assistant/core/issues/103112
urban thorn
i get the same error in my logs but setup is working fine, so this may not be causing the issue.
and different hardware in my case too
have you tried enabling assist_pipeline debugging to listen to the audio being captured?
regal heath
Where do you grab the audio being captured from?
@delicate wadi If you happen to have an accent some people switch to EN-au and get better results.
urban thorn
add the following to configuration.yaml then restart HA with the device disconnected. change dir to a folder that's accessible if you cant access the one below
assist_pipeline:
debug_recording_dir: /share/assist_pipeline
.. as soon as device is connected it will record a wav every 5 seconds regardless of wakeword , so can just talk away. I found easiest way is to plug device in and watch folder , once files start appearing disconnect device. otherwise end up with hundreds of files. also make sure to remove from configuration yaml afterwards otherwise it will just keep creating files.
regal heath
Thank you
delicate wadi
Ok, so it made one 44byte file and isn't making any more
Traceback (most recent call last):
File "/lsiopy/lib/python3.11/site-packages/homeassistant/components/assist_pipeline/pipeline.py", line 1202, in _pipeline_debug_recording_thread_proc
message = queue.get(timeout=message_timeout)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.11/queue.py", line 179, in get
raise Empty
_queue.Empty```It also works if I make it a media player.
topaz fjord
@delicate wadi Did you figure out what's causing the error?
delicate wadi
nope
topaz fjord
I suspect the Atom Echo is not streaming any audio back to HASS
The empty and 44kb wav file implies that
topaz fjord
@delicate wadi Are you running HA in Docker? If so, is it running on the host network (not within a Docker bridge network)?
delicate wadi
I'm running everything in docker. My HA has it's own IP assigned to it on my network via a custom driver (qnap) but it should essentially be like the ipvlan i believe.