Realtime transcription | Nari Labs

Stream 16 kHz PCM16 mono audio as base64 JSON events over an authenticated WebSocket.

Event flow

Receive session.created, send session.configure, then wait for session.configured.
Send input_audio_buffer.append events while receiving optional transcript.partial results.
Send input_audio_buffer.commit. Receive input_audio_buffer.committed, then transcript.completed for that utterance.

VAD and automatic duration boundaries can also commit an utterance. An empty manual commit returns input_audio_buffer.commit_empty. The connection stays open for subsequent utterances.

Stream 16 kHz PCM16 mono audio as base64 JSON events over an authenticated WebSocket. ## Event flow 1. Receive `session.created`, send `session.configure`, then wait for `session.configured`. 2. Send `input_audio_buffer.append` events while receiving optional `transcript.partial` results. 3. Send `input_audio_buffer.commit`. Receive `input_audio_buffer.committed`, then `transcript.completed` for that utterance. VAD and automatic duration boundaries can also commit an utterance. An empty manual commit returns `input_audio_buffer.commit_empty`. The connection stays open for subsequent utterances.

Handshake

WSS

wss://api.narilabs.com/v1/realtime?intent=transcription

Authentication

AuthorizationBearer

Nari API key in the WebSocket handshake Authorization header.

Send

session.configureobjectRequired

Send once within 10 seconds of connecting. Wait for session.configured before sending audio. Open a new connection to change settings.

input_audio_buffer.appendobjectRequired

Send audio in a JSON text frame and receive results concurrently. Appends have no individual acknowledgement.

input_audio_buffer.commitobjectRequired

Commit pending audio. The server returns input_audio_buffer.committed, followed by transcript.completed. With no pending audio, it returns input_audio_buffer.commit_empty instead.

Receive

session.createdobjectRequired

The connection is open. Send session.configure to configure it.

session.configuredobjectRequired

Configuration accepted; audio can now be sent. Returns applied settings and session.limits.

transcript.partialobjectRequired

Replace the previous partial for this item_id; earlier words can change. A completed result can arrive without any partial.

transcript.completedobjectRequired

Final result for this utterance. Replaces any partial text and leaves the connection open.

input_audio_buffer.committedobjectRequired

Acknowledges a manual, VAD, or duration boundary. Use item_id to match the subsequent transcript.completed result.

input_audio_buffer.speech_startedobjectRequired

Sent when server VAD detects the start of speech.

input_audio_buffer.speech_stoppedobjectRequired

Sent when server VAD detects enough silence to end speech.

errorobjectRequired

Reports a failure and closes the connection. See error codes and reconnect handling.

input_audio_buffer.commit_emptyobjectRequired

No audio is pending, so this commit produces no new final result. Previously committed utterances may still have results pending.

URL	wss://api.narilabs.com/v1/realtime?intent=transcription
Method	GET
Status	101 Switching Protocols

Event flow

HandshakeTry it

Authentication

Send

Receive

Event flow

HandshakeTry it

Authentication

Send

Receive

Handshake

Handshake