Update more realtime spec #397
Draft
Conversation
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
- Added `Cancelled` variant to `ResponseStatusDetail` enum for better handling of cancelled responses. - Introduced `LogProb` struct to capture log probability information for transcribed tokens. - Updated `ConversationItemInputAudioTranscriptionCompletedEvent` and `ConversationItemInputAudioTranscriptionDeltaEvent` to include optional `logprobs` for per-token log probability data. - Enhanced `AudioTranscription` struct with optional fields for `language`, `model`, and `prompt` to improve transcription accuracy and customization. - Added new `SemanticVAD` option in the `TurnDetection` enum to control model response eagerness. - Expanded `RealtimeVoice` enum with additional voice options for more variety in audio responses.
- Changed enum variants for `AudioFormat` to use underscores instead of hyphens in their serialized names. - Updated `G711ULAW` from `g711-ulaw` to `g711_law` and `G711ALAW` from `g711-alaw` to `g711_alaw` for improved clarity and adherence to naming conventions.
…nd response management
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Outdated suggestions cannot be applied.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.
🚀 Summary
Syncs
async-openairealtime types with the latest OpenAI Realtime API (June 2025).Adds richer request/response configs, new client & server events, extra enums for models / voices / modalities, plus tracing & noise-reduction support.
✨ What’s new
Client events
ResponseConfig,OutputAudioBufferClearEvent,ConversationItemRetrieveEvent.ResponseCancelEventgainsresponse_id.ResponseCreateEventnow usesResponseConfiginstead ofSessionResource.Server events
output_audio_buffer.cleared,conversation.item.input_audio_transcription.delta,conversation.item.retrieved.InputAudioBufferCommitedEvent→InputAudioBufferCommittedEvent.Response resource
finish_reason,created_at.TokenLimit,FunctionCall.Session resource
RealtimeModel,Modality,NoiseReductionType.speed,input_audio_noise_reduction,tracing.modelis nowRealtimeModel;modalitiesisVec<Modality>.Turn detection
semantic_vadmode withcreate_responseandinterrupt_responseflags.Audio
g711_ulaw,g711_alaw).InputAudioNoiseReduction.Tooling
ToolChoice&ToolDefinitionintoResponseConfig.ResponseCreateEvent:responsenow expectsResponseConfig, notSessionResource.g711-ulaw/g711-alaw→g711_ulaw/g711_alaw.InputAudioBufferCommitedEvent→InputAudioBufferCommittedEvent.SessionResource.modelis nowRealtimeModel(no longer a free-formString).