feat(assemblyai): add universal-3-5-pro (now default) and Voice Focus streaming params#6119
Open
dlange-aai wants to merge 1 commit into
Open
feat(assemblyai): add universal-3-5-pro (now default) and Voice Focus streaming params#6119dlange-aai wants to merge 1 commit into
dlange-aai wants to merge 1 commit into
Conversation
… params
Add universal-3-5-pro as a u3-rt-pro-family streaming model and make it the
plugin default. It inherits the full u3-rt-pro parameter set (prompt,
agent_context, previous_context_n_turns, continuous_partials,
interruption_delay) and the family connect-time defaults.
Add Voice Focus support via the voice_focus ("near-field" / "far-field") and
voice_focus_threshold connection parameters, gated to the u3-rt-pro family and
set at connect time only.
Consolidate the per-parameter u3-rt-pro-family validation into a single loop so
the error message lists the actual supported models (it previously named only
'u3-rt-pro', now misleading since universal-3-5-pro is the default).
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
universal-streaming-english). It joins the u3-rt-pro parameter family, so it accepts the full u3-pro parameter set (prompt,agent_context,previous_context_n_turns,continuous_partials,interruption_delay) and inherits the family's connect-time defaults (continuous_partials=True,language_detection=True, punctuation-based turn detection)."near-field"for headsets/handsets/close-talking mics,"far-field"for conference rooms/laptop mics. Isolates the primary voice and suppresses background noise before audio reaches the model. See https://www.assemblyai.com/docs/streaming/voice-focus.[0.0, 1.0]controlling how aggressively background audio is suppressed.ValueErrorlists the actual supported models (it previously named onlyu3-rt-pro, now misleading sinceuniversal-3-5-prois the default).Test Plan
uv run pytest --plugin assemblyai— 51 tests (default-model assertions, u3-rt-pro-family gating + acceptance across all family models,voice_focus/voice_focus_thresholdplumbing, connect-config forwarding incl.voice_focus_threshold=0.0,update_optionsexclusion). 5 existing gating tests were updated to pinmodel="universal-streaming-english"since the default now joins the family.