---
title: "Create transcription session"
method: POST
path: "/api/v1/transcription/session/create"
tags: ["/api/v1/transcription/session"]
---

# Create transcription session

`POST /api/v1/transcription/session/create`

Creates a transcription session for real-time speech-to-text over WebSocket. Returns a `transcription_session_id` to use when connecting to `/ws/transcribe`.

## Headers

- `sdp_provider_id` string

## Request body

- ControllersCreateTranscriptionSessionRequest — Optional transcription session ID and audio configuration. Suki generates an ID when omitted.
  - `audio_config` ControllersTranscriptionAudioConfig — Audio configuration for transcription
    - `audio_encoding` 'LINEAR16' — **Optional** - Audio encoding format
    - `audio_language` string — **Optional** - Language code for audio transcription
    - `sample_rate_hertz` integer — **Optional** - Sample rate in Hertz
  - `transcription_session_id` string — **Optional** - UUID format. If not provided, a session ID will be automatically generated.

## Response `201`

Resource created successfully.

- ControllersCreateTranscriptionSessionResponse — Response body for the /transcription/session/create endpoint
  - `transcription_session_id` string — Unique identifier for the transcription session

## Other responses

- `400` — Bad request. The request body or parameters failed validation.
- `401` — Unauthorized. The Suki access token is missing, expired, or invalid.
- `403` — Forbidden. The authenticated user cannot access this resource.
- `404` — Not found. The session, encounter, or resource ID does not exist.
- `500` — Internal server error.

---

[API](https://skmtc.dev/suki/apis/suki-developer-platform.md) · [All operations](https://skmtc.dev/suki/apis/suki-developer-platform/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/suki/suki-developer-platform/revisions/5f420a18fb68/schema)