fix(realtime): preserve output_audio content parts in output_item events#3230
Merged
seratch merged 1 commit intoopenai:mainfrom May 9, 2026
Merged
Conversation
The dict-based fast path in _handle_ws_event for response.output_item.added and response.output_item.done only matched part type "audio". GA-style assistant messages emit "output_audio" (matching the official Content type literal in openai.types.realtime), so audio content + transcript were silently dropped from listeners. Accept both "audio" and "output_audio" and normalize to the SDK's "audio" type, mirroring how output_text/text are already handled in the same block.
Member
|
@codex review |
|
Codex Review: Didn't find any major issues. Another round soon, please! ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. Codex can also answer questions or update the PR. Try commenting "@codex address that feedback". |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 dict-based fast path in
_handle_ws_eventforresponse.output_item.added/response.output_item.doneonly matched part type"audio". GA-style assistant messages emit"output_audio"(matching the official literalLiteral["output_text", "output_audio"]onopenai.types.realtime.realtime_conversation_item_assistant_message.Content), so audio content + transcript were silently dropped from listeners on these events.The other early conversion site (
_ConversionHelper.conversation_item_to_realtime_message_itemat L1588) already mapsoutput_audio→audio; the dict-path was inconsistent.This PR accepts both
"audio"and"output_audio"and normalizes to the SDK's"audio"type, mirroring howoutput_text/textare already handled in the same block.Changes
src/agents/realtime/openai_realtime.py: 1-line tuple change in_handle_ws_event's message-content normalization.tests/realtime/test_openai_realtime.py: focused regression test assertingoutput_audioparts reach listeners withtype="audio"and the transcript preserved.Diff: 2 files, +35/-1.
Test plan
test_output_audio_content_type_normalizedfails onmain, passes after fix.tests/realtime/suite (233 tests) passes.ruff checkclean.