Intend-failed : Unexpected error during intent recognition

Hi,

I’m trying to setup a local assist pipeline and I’m running into a strange issue.
I tried searching a bit on the web for this issue and all I could find are related to malformed custom_sentences files. I do not have a custom sentence file (yet?). I’m at the very beginning of my local voice assistant journey.

Here’s what I currently have configured that relates to speech handling (TTS and STT):

  • Google assistant manually configured (ie no through Nabu Casa) and it’s been working great.

  • I have a few alerts triggering Microsoft TTS notifications through my Google Home enabled sound bar.

Here’s an example of the raw output of the assist pipeline when I try to input a command.

stage: done
run:
  pipeline: 01gzv2ccyq72ebaj4h4wk2j58q
  language: fr
  runner_data:
    stt_binary_handler_id: 1
    timeout: 300
events:
  - type: run-start
    data:
      pipeline: 01gzv2ccyq72ebaj4h4wk2j58q
      language: fr
      runner_data:
        stt_binary_handler_id: 1
        timeout: 300
    timestamp: "2024-01-30T13:57:19.724304+00:00"
  - type: stt-start
    data:
      engine: stt.faster_whisper
      metadata:
        language: fr
        format: wav
        codec: pcm
        bit_rate: 16
        sample_rate: 16000
        channel: 1
    timestamp: "2024-01-30T13:57:19.724535+00:00"
  - type: stt-vad-start
    data:
      timestamp: 455
    timestamp: "2024-01-30T13:57:20.751608+00:00"
  - type: stt-vad-end
    data:
      timestamp: 1265
    timestamp: "2024-01-30T13:57:22.351438+00:00"
  - type: stt-end
    data:
      stt_output:
        text: " Allumer le lumière chambre"
    timestamp: "2024-01-30T13:57:22.777072+00:00"
  - type: intent-start
    data:
      engine: homeassistant
      language: fr-CA
      intent_input: " Allumer le lumière chambre"
      conversation_id: null
      device_id: null
    timestamp: "2024-01-30T13:57:22.777119+00:00"
  - type: error
    data:
      code: intent-failed
      message: Unexpected error during intent recognition
    timestamp: "2024-01-30T13:57:22.869953+00:00"
  - type: run-end
    data: null
    timestamp: "2024-01-30T13:57:22.870000+00:00"
stt:
  engine: stt.faster_whisper
  metadata:
    language: fr
    format: wav
    codec: pcm
    bit_rate: 16
    sample_rate: 16000
    channel: 1
  done: true
  stt_output:
    text: " Allumer le lumière chambre"
intent:
  engine: homeassistant
  language: fr-CA
  intent_input: " Allumer le lumière chambre"
  conversation_id: null
  device_id: null
  done: false
error:
  code: intent-failed
  message: Unexpected error during intent recognition

Here’s an error log generated when running assist pipeline

Logger: homeassistant.components.assist_pipeline.pipeline
Source: components/assist_pipeline/pipeline.py:988
Integration: Assist pipeline (documentation, issues)
First occurred: 8:57:22 AM (1 occurrences)
Last logged: 8:57:22 AM

Unexpected error during intent recognition
Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/components/assist_pipeline/pipeline.py", line 988, in recognize_intent
    conversation_result = await conversation.async_converse(
                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/homeassistant/homeassistant/components/conversation/__init__.py", line 467, in async_converse
    result = await agent.async_process(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/homeassistant/homeassistant/components/conversation/default_agent.py", line 211, in async_process
    result = await self.async_recognize(user_input)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/homeassistant/homeassistant/components/conversation/default_agent.py", line 193, in async_recognize
    result = await self.hass.async_add_executor_job(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/concurrent/futures/thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/homeassistant/homeassistant/components/conversation/default_agent.py", line 290, in _recognize
    for result in recognize_all(
  File "/usr/local/lib/python3.11/site-packages/hassil/recognize.py", line 453, in recognize_all
    for maybe_match_context in maybe_match_contexts:
  File "/usr/local/lib/python3.11/site-packages/hassil/recognize.py", line 925, in match_expression
    group_contexts = [
                     ^
  File "/usr/local/lib/python3.11/site-packages/hassil/recognize.py", line 925, in <listcomp>
    group_contexts = [
                     ^
  File "/usr/local/lib/python3.11/site-packages/hassil/recognize.py", line 917, in match_expression
    yield from match_expression(settings, context, item)
  File "/usr/local/lib/python3.11/site-packages/hassil/recognize.py", line 925, in match_expression
    group_contexts = [
                     ^
  File "/usr/local/lib/python3.11/site-packages/hassil/recognize.py", line 925, in <listcomp>
    group_contexts = [
                     ^
  File "/usr/local/lib/python3.11/site-packages/hassil/recognize.py", line 1176, in match_expression
    raise MissingRuleError(f"Missing expansion rule <{rule_ref.rule_name}>")
hassil.recognize.MissingRuleError: Missing expansion rule <quelest>

Whisper addon is configured and running succesfully (wav file is being mostly correctly interpreted).
I currently disabled any TTS engine to try and limit any source of issues.

I’m running a supervised installed on Debian.