Skip to main content
The Dictation APIs enable partners to run speech-to-text without the full ambient clinical note flow. You create a transcription session, open a WebSocket to stream audio, then end the session when capture is finished so resources close cleanly. These APIs are called from your servers and authenticated with a Suki Token (sdp_suki_token). Critically, Dictation returns transcript text for the session. It does not replace Ambient Session Management when you need a generated clinical note.

​
Available endpoints

Create Dictation Session

Create a Dictation session and receive a transcription session ID

Stream Dictation Session

Stream audio over WebSocket and receive transcript frames

End Dictation Session

End the Dictation session when capture is finished

Dictation Streaming

Learn how to stream Dictation audio and receive transcript frames

Audio Dictation Overview

Learn how Dictation works and the different modes it supports

WebSocket Streaming Wire Format

Review outbound audio message framing for Dictation WebSockets

Read Dictation Transcript Frames

Learn how to parse inbound transcript frames from the Dictation WebSocket

​
Common use cases

Dictate into chart fields and scratchpads

Capture speech-to-text for the focused field or scratchpad without running the full ambient clinical note pipeline.

Correct ambient drafts in place

Let clinicians fix wording in a generated note with Dictation, then save the edited content in your existing chart workflow.

Stream transcripts on custom clients

Own audio capture on a backend or device gateway while your UI renders live transcript results for the clinician.

Support push-to-talk dictation

Keep one Dictation session available across multiple speech bursts until the clinician finishes the documentation task.
Last modified on July 24, 2026