Types

Type aliases, union types, and utility types.

• AgentEvent — Union of all agent lifecycle events.
• AgentState — The possible states of the CompositeVoice agent.
• AudioCaptureState — Runtime state of the audio capture pipeline.
• AudioEncoding — Audio encoding schemes supported by the SDK.
• AudioEvent — Union of all audio-related events.
• AudioFormat — Audio container/codec formats supported by the SDK.
• AudioPlaybackState — Runtime state of the audio playback pipeline.
• AwsCredentialsProvider — AWS credentials as a static object or an async factory.
• AzureSTTRecognitionMode — Azure Speech recognition modes.
• AzureTTSOutputFormat — Supported Azure Speech TTS output formats.
• BlocklistAction — What a blocklist match does to the utterance.
• CartesiaOutputEncoding — Cartesia output audio encoding types.
• CartesiaTTSModel — Cartesia voice model identifiers.
• CompositeVoiceEvent — Union of all events that can be emitted by a CompositeVoice agent.
• DeepgramAgentEvent — Discriminated union of events emitted by the Deepgram Agent provider.
• DetectedAudioFormat — Audio container/codec formats detectable via magic byte inspection.
• DrainCallback() — Callback type for the drain consumer.
• ElevenLabsAgentEvent — Discriminated union of events emitted by ElevenLabsAgent via onAgentEvent.
• ElevenLabsOutputFormat — ElevenLabs output format identifiers.
• ElevenLabsSTTAudioFormat — ElevenLabs STT audio format identifiers.
• ElevenLabsSTTModel — ElevenLabs STT model identifiers.
• ElevenLabsTTSModel — ElevenLabs voice model identifiers.
• EventListener()— Generic event listener function type.
• EventStreamHeaderValue — A decoded event-stream header value.
• EventType — String union of all possible event type identifiers.
• FishAudioTTSFormat — Supported Fish Audio TTS audio output formats.
• FishAudioTTSLatency — Latency mode for Fish Audio synthesis.
• FishAudioTTSModel — Supported Fish Audio TTS model generations.
• GeminiLiveAgentEvent — Discriminated union of events emitted by GeminiLiveAgent via onAgentEvent.
• GladiaSTTEncoding — Audio encodings accepted by the Gladia live API.
• GladiaSTTRegion — Gladia processing regions.
• GoogleMeetConnectionState — Connection state of the Meet media session, as reported on the session-control data channel.
• GoogleSTTEncoding — Supported Google Cloud STT (v1) audio encodings.
• GoogleTTSAudioEncoding — Supported Google Cloud TTS audio encodings.
• GoogleTTSSsmlGender — SSML voice gender preference for Google Cloud TTS voice selection.
• GuardrailErrorPolicy — What to do when a guardrail throws or exceeds its timeout.
• GuardrailEvent — Union of all guardrail-related events.
• GuardrailMode — When guardrails run relative to TTS streaming.
• GuardrailSegmentation — How streaming text is cut into units before guardrails see it.
• GuardrailStage — Pipeline point at which a guardrail runs.
• InputAttachTarget— Extracts the attach() target type declared by an AttachableInputProvider in a providers array.
• LLMEvent — Union of all LLM-related events.
• LLMStreamChunk — Discriminated union for streaming chunks from tool-aware generation.
• LMNTTTSFormat — Supported LMNT TTS audio output formats.
• LMNTTTSModel — Supported LMNT TTS models.
• LMNTTTSSampleRate — Supported LMNT TTS output sample rates in Hz.
• MiniMaxTTSEmotion — Supported MiniMax TTS voice emotions.
• MiniMaxTTSFormat — Supported MiniMax TTS audio output formats.
• MiniMaxTTSModel — Supported MiniMax TTS models.
• ModerationFn() — Classifies text as acceptable or not.
• MurfTTSChannelType — Supported Murf TTS channel types.
• MurfTTSFormat — Supported Murf TTS audio output formats.
• MurfTTSModelVersion — Supported Murf TTS model versions.
• MurfTTSSampleRate — Supported Murf TTS output sample rates in Hz.
• OpenAIRealtimeAgentEvent — Discriminated union of events emitted by OpenAIRealtimeAgent via onAgentEvent.
• OpenAIRealtimeTurnDetection — Turn-detection configuration for OpenAI Realtime transcription sessions.
• OpenAITTSFormat — Supported OpenAI TTS audio output formats.
• OpenAITTSVoice — Supported OpenAI TTS voice identifiers.
• OverflowCallback() — Callback type for overflow notifications.
• PIIType — Categories of personally identifiable information recognized by createPIIRedactionGuardrail.
• PollyEngine — Amazon Polly synthesis engines.
• PollyOutputFormat — Supported Amazon Polly audio output formats.
• ProviderEvent — Union of all provider-level events.
• ProviderFallbackReason — Why a fallback chain swapped to another provider.
• ProviderRole — Pipeline role a provider can fulfil.
• ProviderType — Communication type for providers.
• QueueEvent — Union of all queue-related events.
• ReconcileResult — Result of reconciling a preflight with confirmed text.
• RimeTTSFormat — Supported Rime TTS audio output formats.
• RimeTTSModel — Supported Rime TTS models.
• SmallestTTSFormat — Supported Smallest.ai Waves TTS audio output formats.
• SmallestTTSModel — Supported Smallest.ai Waves TTS models.
• SmallestTTSSampleRate — Supported output sample rates (Hz) for the Waves TTS API.
• SpeakProvider — Union of all supported TTS provider configurations for the Agent API.
• SpeechifyTTSFormat — Supported Speechify TTS audio output formats.
• SpeechifyTTSModel — Supported Speechify TTS models.
• StartListeningArgs— Parameter tuple for CompositeVoice.startListening(), derived from the configured providers.
• STTProvider — Union type for all speech-to-text providers.
• TeamsCallState — Call states reported by the ACS Calling SDK’s Call.state.
• ThinkProvider — Union of all supported LLM provider configurations for the Agent API.
• TranscribePartialResultsStability — Stability levels for Amazon Transcribe partial-results stabilization.
• TranscriptionEvent — Union of all transcription-related events.
• TTSEvent — Union of all TTS-related events.
• TTSProvider — Union type for all text-to-speech providers.
• TurnEvent — Union of all turn-level events.
• TwilioSocketListener() — Loosest-possible event-listener shape for TwilioStreamSocket.
• VADEvent — Union of all local-VAD events.
• VonageSocketListener() — Loosest-possible event-listener shape for VonageSocket.
• ZoomRtmsDataOption — Audio stream layout requested from RTMS.
• ZoomRtmsSampleRate — Audio sample rates supported by the RTMS media handshake.