Retrieve Results

Get a completed transcript or caption document.

Endpoint

GET /api/v3/speech-to-text/session/{session_id}/document/{document_id}

Path Parameters

Parameter	Type	Description
session_id	string	Session ID
document_id	string	Document ID

Query Parameters

Parameter	Type	Description
format	string	Export format (optional)
template_id	string	Export template ID (optional)

Response Type

Response Format: BaseResponse<DocumentObject>

{
  "count": 1,
  "total_results": 1,
  "result": {
    "id": "document_id",
    "filename": "example.mp4",
    "type": "document",
    "language": "en",
    "created": "2025-01-15T10:30:00Z",
    "last_modified": "2025-01-15T10:45:00Z",
    "finished": true,
    "content_type": "application/json",
    "document": {
      "document_type": "document",
      "version": "1.0",
      "document": []
    }
  }
}

Document Object Fields

Field	Type	Description
id	string	Document ID (readonly)
filename	string	Original filename (readonly)
type	string	Document type
language	string	Language code
created	Date	Creation timestamp (readonly)
last_modified	Date	Last modification timestamp (readonly)
finished	boolean	Whether document is marked as finished
timecode_offset	string	Timecode offset (optional)
document	object	Document content (null when direct_download=true)
download_url	string | null	Direct download URL (when direct_download=true)
expires_in	number | null	URL expiration in seconds (when direct_download=true)
content_type	string	Content type
size_bytes	number | null	File size in bytes

Document Structure

For transcript documents (document_type: "document"):

The document field contains a DocumentTranscriptV1 object with:

• document_type: "document"
• version: "1.0"
• document: Array of utterances

Utterance Structure:

• type: "utterance"
• speaker: Speaker identifier (if diarization enabled)
• start: Start time in milliseconds
• stop: Stop time in milliseconds
• children: Array of paragraphs

Paragraph Structure:

• type: "paragraph"
• start: Start time in milliseconds
• stop: Stop time in milliseconds
• children: Array of text objects

Text Object Structure:

• text: Text content
• start: Start time in milliseconds
• stop: Stop time in milliseconds
• bold: Boolean (optional)
• italic: Boolean (optional)
• underlined: Boolean (optional)
• strike: Boolean (optional)
• color: String (optional)
• mark: Boolean (optional)

Shared Document Endpoint

For shared documents (token-based access without authentication):

GET /api/v3/speech-to-text/session/{session_id}/shared/document/{document_id}

See Documents API.

Related

• Direct Blob Upload
• Check Status
• Documents API