Initial multiwake support #264
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
stintel
force-pushed
the
feature/multiwake
branch
3 times, most recently
from
8dd390c
to
a31ac0c
Compare
This requires modifications to ESP-ADF, so we need to bump it to a newer version.
We currently do LVGL calls when handling AUDIO_REC_WAKEUP_END and MSG_STOP. We trigger MSG_STOP from AUDIO_REC_WAKEUP_END, so it's better to consolidate the LVGL calls so we only need to acquire a lock once. Move the LVGL call from AUDIO_REC_WAKEUP_END to MSG_STOP.
The wake_start message will send the wake audio volume to WAS, needed to implement multiwake support. When the WAS WebSocket is disconnected, skip sending the message, as this is latency sensitive, and re-establishing the connection would take too long. By then, the wake session would have already expired. Another problem with trying to reconnect on wake, is that when WAS is still down, the connection attempt times out and slows down the entire operation of wake, vad, stream, ... and Willow appears to hang. Finally, Willow will try to reconnect to WAS anyway.
We cannot write the "WOW Active - Exiting" message from the WAS code, as it would be overwritten with "Thinking" on AUDIO_REC_WAKEUP_END, which is triggered by audio_recorder_trigger_stop.
When VAD_START is triggered before we got the multiwake result from WAS, we have started streaming to WIS already. Use the multiwake_won boolean to bail out early when handling the HTTP_STREAM_FINISH_REQUEST event, to avoid updating the UI or playing chimes or TTS when we lost the multiwake race. Ideally we only start streaming to WIS once we get the wake result, but this will require some refactoring.
stintel
force-pushed
the
feature/multiwake
branch
from
4225c91
to
abfc331
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR implements handling of multiple Willow devices waking at the same time. With WAS support, only the Willow device with the highest wake word volume will send a command to the endpoint. Without WAS support, Willow will function like before.
It is possible to trigger VAD START before the wake_result is received from WAS, which will cause Willow devices who have lost the wake race to start streaming to WIS already. This should be improved, but will require some refactoring. For now, this should be good enough to demonstrate the feature, but the feature should be considered experimental.