Unexpected error during intent recognition in 2024.10.0b0

[caohuongls]

I've upgraded to the new beta version 2024.10.0b0 for testing, but I've found that this custom component not working with error " Unexpected error during intent recognition". However, when others agent runs independently, they work fine.

This occurs in the combinations between the Assist, Extended OpenAI Conversation, and Google Generative AI conversation agents.

``

• type: error data: code: intent-failed message: Unexpected error during intent recognition timestamp: "2024-09-26T10:18:28.629196+00:00" `` 2024-09-26 17:43:11.310 ERROR (MainThread) [homeassistant.components.assist_pipeline.pipeline] Unexpected error during intent recognition Traceback (most recent call last): File "/usr/src/homeassistant/homeassistant/components/assist_pipeline/pipeline.py", line 1017, in recognize_intent conversation_result = await conversation.async_converse( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/src/homeassistant/homeassistant/components/conversation/agent_manager.py", line 110, in async_converse result = await method(conversation_input) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/src/homeassistant/homeassistant/components/conversation/entity.py", line 47, in internal_async_process return await self.async_process(user_input) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/config/custom_components/fallback_conversation/conversation.py", line 96, in async_process default_agent = conversation.default_agent.async_get_default_agent(self.hass) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: module 'homeassistant.components.conversation.default_agent' has no attribute 'async_get_default_agent'. Did you mean: 'async_setup_default_agent'?

[m50]

Can you upgrade Fallback Conversation Agent to v2.0.2-beta and see if it fixes the issue? I am not on the beta build of HomeAssistant, and I prefer stability, so never upgrade until a couple weeks after release, and do not have a beta HA setup (I should probably do that, considering they keep breaking things) so I won't be able to test locally right now

[caohuongls]

@m50 Thanks, it seems to have fixed the issue for me. Everything is working as expected.

[caohuongls]

I have another question that's not directly related to this issue. I hope you have experience with it, and we can discuss it further in a different context. I'm using Wyoming Satellite on a Raspberry Pi 2W with your component to control Home Assistant. The issue I'm facing is:

• With the Atom Echo M5, the conversation flows continuously and feels very natural.

• However, on the Wyoming Satellite device, after completing the first response, I have to trigger the wake word again and This causes Home Assistant or Wyoming Satellite to start a new conversation.

Have you encountered this issue? Is it something that can be fixed within your component, or is it on the Wyoming Satellite side? Thanks

[m50]

It's a wyoming Satellite limitation. Something I wanted to try to fix in it's code, but it seems like the HA team is moving away from wyoming satellite so I gave up on that. But yeah, not related to this at all.

[iganeshk]

@m50 just checking in to see if you were aware of this PR https://github.com/home-assistant/core/pull/118835 which should have addressed this issue.

I am trying to back trace this issue still persists as this component should pick up the conversation_id by design https://github.com/m50/ha-fallback-conversation/blob/523d91ac31bf9a33119985a8f17fda84b78c26e9/custom_components/fallback_conversation/conversation.py#L113-L116

Edit: Looks like the Satellite Source itself has this debt, sorry to open this back up!

[maxi1134]

I have this issue on v2.0.2 coupled with HA2024.10.1.

Is anyone else experiencing it?