Skip to main content
Audio Dictation converts provider speech into text while a session is active. Providers can dictate during a visit or into a field in your application while your product receives partial and final transcript text. Audio Dictation is different from Clinical documentation (ambient), which transcribes the visit and generates structured clinical notes, LOINC-based sections, and related outputs. Audio Dictation gives you speech-to-text you can route into your own UI, EHR fields, or downstream workflows.

​
When to use Audio Dictation

Choose Audio Dictation when you need:
  • Real-time transcription without generating a full ambient clinical note.
  • Transcript text you control in your application (display, edit, save, or send elsewhere).
  • Browser-based Dictation with Suki-managed UI (Web SDK or Dictation SDK), or full control with REST and WebSocket APIs.
Choose Ambient clinical documentation when you need visit capture, note generation, structured clinical data, and related ambient outputs.

​
Developer workflow

The Audio Dictation developer workflow is shared by Dictation APIs, Web SDK, and Dictation SDK. The only difference is who owns each step.
1

Start Dictation

Start a Dictation session for a field, note section, or scratchpad in your UI. With Partner APIs, create the session over REST. With Web SDK or Dictation SDK, start Dictation through the SDK controls or hosted experience.
2

Capture Speech

Capture provider speech while Dictation is active. Partner APIs stream audio over WebSocket on GET /ws/transcribe. Web SDK and Dictation SDK handle browser capture for you.
3

Receive Transcripts

Receive partial and final transcript text as the provider speaks. Partner APIs deliver transcript frames on the WebSocket. SDKs deliver text through events or callbacks.
4

Insert Text

Insert the transcript into the active target in your application. Your product owns field focus, rendering, and where the text lands.
5

Review and Save

Let the provider review and correct the text, then save it in your application or EHR workflow.
The steps above stay the same across products. What changes is how much you build yourself. With Dictation APIs, your application creates the session, streams audio, and handles transcript frames. With Web SDK or Dictation SDK, Suki handles more of capture and UI. In every case, your application still decides where final text lands and how you save it. If you are building with Partner APIs, follow Dictation basic usage. To choose a modality and design the clinician experience, see the Audio Dictation guide.

Audio Dictation Guide

Capabilities, common use cases, how Dictation works, and how to choose APIs, Web SDK, or Dictation SDK.

Audio Dictation API Basic Usage

Step-by-step Partner API workflow: create a session, stream audio, receive transcripts, and end the session.

Dictation Audio Streaming

How to stream Dictation audio on GET /ws/transcribe, with links to wire format and transcript frames.
Refer to these guides when you already know your integration path and want to dive deeper:

Dictation API References

REST and WebSocket references for transcription sessions.

Web SDK for Audio Dictation

JavaScript and React libraries with Suki Web SDK packages.

Dictation SDK

Hosted iframe with Suki Dictation SDK packages.
This SDK is currently in Beta and is not yet generally available.

​
Next steps

Complete the Partner onboarding process and get your credentials.
Last modified on August 13, 2026