Amazon Q Business will no longer be open to new customers starting on July 31, 2026. If you would like to use the service, please sign up prior to July 30. For capabilities similar to Q Business, explore Amazon Quick. Learn more.

GetDocumentContent Output Schema

When you use the GetDocumentContent API with outputFormat set to EXTRACTED, the response returns extracted text content in JSON format. The output schema is presented in JSON format:

{
   // always V1 for now
   schemaVersionId: string;
   
   // always JSON for now
   outputFormat: string;
   
   // content for plain-text documents
   plainTextDocumentContent: string;
   
   // content for non-plaintext documents such as PDF, DOCX, PPTX, Audio, Video
   nonPlainTextDocumentContent: List<ExtractedDocumentBodyElement>;
}

The schema for non-plaintext documents includes the ExtractedDocumentBodyElement which includes:

{
   text: string;
   
  // Allowed values: TEXT, ARTICLE, SECTION, DIV, IMAGE_DESCRIPTION, CODE,
  // TABLE, LIST, URL, HEADER, FOOTER, FORM, MENU, AUDIO, VIDEO
   elementType: string;
   
   horizontalHeaderIndex: integer;
   verticalHeaderIndex: integer;
   htmlDocumentTitle: string;
   sectionTitle: string;
   sectionBody: string;
   tableCaption: string;
   tableFooter: string;
   tableRowHeaders: List<List<string>>;
   tableColumnHeaders: List<List<string>>;
   tableRows: List<List<string>>;
   tableRowsCount: integer;
   tableColumnsCount: integer;
   tableId: string;
   
   tokens: List<struct>;
   {
        value: string;
        startOffsets: integer;
        endOffsets: integer;
   }
   
   tableType: string;
   tableSummary: string;
   
   columnInfoList: List<struct>;
   {
        columnName: string;
        columnSummary: string;
        columnType: string;
        columnRepresentativeValues: List<string>
   }
   
   // Audio/Video specific fields below
   overallSummary: string;
   
   audioSummaryList: List<struct>;
   {
        summaryText: string;
        startTimeMilliseconds: string;
        endTimeMilliseconds: string;
   }
   
   videoSummaryList: List<struct>;
   {
        summaryText: string;
        startTimeMilliseconds: string;
        endTimeMilliseconds: string;
   }
   
   audioTranscriptList: List<struct>;
   {
        transcriptText: string;
        startTimeMilliseconds: string;
        endTimeMilliseconds: string;
   }
   
   videoTranscriptList: List<struct>;
   {
        transcriptText: string;
        startTimeMilliseconds: string;
        endTimeMilliseconds: string;
   }   
}

Example Output

Plaintext Document Example

For plaintext documents, the extracted content is returned in the plainTextDocumentContent field:

{
  "schemaVersionId": "V1",
  "outputFormat": "JSON",
  "plainTextDocumentContent": "This is the extracted text content from a plain text document."
}