Quickstart - Transcription

Speech to text from a pre-recorded audio file

Introduction

The Transcription Workflow is defined by processing a complete audio file to return a text document. In scenarios where real-time speech-to-text is not required or feasible, the transcription workflow provides functional and cost effective means for creating verbatim, conversational or dictation-style transcripts.

Endpoints and capabilities

Endpoint	Capability	Use
Interactions	The foundational unit that ties together all related data and operations, enabling a cohesive workflow.	Required
Recordings	Upload audio file(s) that can be used for transcript generation.	Required
Transcripts	Generate transcripts for audio files that are associated with the interaction.	Required

See audio format requirements here

---

Workflow

1. Create interaction

   1. The workflow begins with the client initiating an interaction by sending a POST request to the /interactions endpoint.
   2. The API responds with a unique interactionId for the interaction and a WebSocket URL (websocketUrl). The identifier will be used to manage the subsequent steps of the workflow. The WebSocket URL will not be required for this workflow.
2. Upload audio recording

   3. Once the interaction is initialized, the client uploads an audio file associated with that interaction by sending a POST request to /interactions/:id/recordings.
   4. The API responds with a 201 status and returns a recordingId, confirming that the audio file has been successfully uploaded and linked to the interaction.
3. Create transcript

   5. After the recording is uploaded, the client initiates the transcription process by sending a POST request to /interactions/:id/transcripts.
   6. The API processes the audio and returns a 201 status with the generated transcript. This transcript contains the text version of the recorded interaction, extracted and formatted for review.

See details on transcription configuration options here