Realtime
Realtime speech-to-text transcription service. This WebSocket API enables streaming audio input and receiving transcription results.
## Event Flow
- Audio chunks are sent as `input_audio_chunk` messages
- Transcription results are streamed back in various formats (partial, committed, with timestamps)
- Supports manual commit or VAD-based automatic commit strategies
Authentication is done either by providing a valid API key in the `xi-api-key` header or by providing a valid token in the `token` query parameter. Tokens can be generated from the [single use token endpoint](/docs/api-reference/tokens/create). Use tokens if you want to transcribe audio from the client side.
Handshake
WSS
/v1/speech-to-text/realtime
Headers
xi-api-key
Query parameters
model_id
ID of the model to use for transcription.
token
Single use token for authentication. Only used when initiating a session from the client. If provided, xi-api-key is no longer required for authentication.
include_timestamps
Whether to receive the committed_transcript_with_timestamps event, which includes word-level timestamps.
include_language_detection
Whether to include the detected language code in the committed_transcript_with_timestamps event.
audio_format
Audio encoding format for speech-to-text.
language_code
Language code in ISO 639-1 or ISO 639-3 format.
commit_strategy
Strategy for committing transcriptions.
Allowed values:
vad_silence_threshold_secs
Silence threshold in seconds for VAD.
vad_threshold
Threshold for voice activity detection.
min_speech_duration_ms
Minimum speech duration in milliseconds.
min_silence_duration_ms
Minimum silence duration in milliseconds.
enable_logging
When enable_logging is set to false zero retention mode will be used for the request. This will mean history features are unavailable for this request. Zero retention mode may only be used by enterprise customers.
Send
publish
Receive
Session Started Payload
OR
Partial Transcript Payload
OR
Committed Transcript Payload
OR
Committed Transcript with Timestamps Payload
OR
Scribe Error Payload
OR
Scribe Auth Error Payload
OR
Scribe Quota Exceeded Error Payload
OR
Scribe Throttled Error Payload
OR
Scribe Unaccepted Terms Error Payload
OR
Scribe Rate Limited Error Payload
OR
Scribe Queue Overflow Error Payload
OR
Scribe Resource Exhausted Error Payload
OR
Scribe Session Time Limit Exceeded Error Payload
OR
Scribe Input Error Payload
OR
Scribe Chunk Size Exceeded Error Payload
OR
Scribe Insufficient Audio Activity Error Payload
OR
Scribe Transcriber Error Payload
