fix(assemblyai): stop forcing continuous_partials=true for U3-Pro models#6401
Merged
Conversation
AssemblyAI now enables continuous_partials server-side by default, and deliberately disables it when speaker_labels is on so turns break cleanly at speaker changes. The plugin's forced true default was redundant in the common case and overrode the diarization-tuned behavior. Leave the parameter unset unless explicitly provided.
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
toubatbrian
approved these changes
Jul 13, 2026
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
The plugin defaulted
continuous_partials=Truefor U3-Pro family models, added in #5819 when AssemblyAI's server default wasfalse. AssemblyAI has since flipped the server default to enabled (docs), making the plugin override redundant in the common case — and actively counterproductive whenspeaker_labelsis enabled, where the server deliberately disables continuous partials (along with diarization-tuned silence values) so turns break cleanly at speaker changes, only to have the plugin's explicitcontinuous_partials=truestomp it.Changes
continuous_partials=Truedefault; an unset value staysNOT_GIVENand is not sent on the wire, so AssemblyAI's server defaults govern (enabled normally, disabled whenspeaker_labelsis on). Explicit user-passed values are unaffected, including viaupdate_options/ mid-sessionUpdateConfiguration.continuous_partialsdocstring to describe the current server defaults.Not addressed here: the plugin's
min_turn_silence=100latency default similarly overrides the server's diarization-tuned 640/768 ms values whenspeaker_labelsis on — left as a separate decision since it's an intentional latency choice.