Disable OCR in Adobe PDF Services When Converting Scanned PDFs to DOCX

Report · Oct 28, 2024

I'm using the Adobe PDF Services API to convert PDF files to DOCX format.

My setup works well, but I’m encountering an issue with scanned documents. (net8.0, C#)

When I upload a scanned PDF, the service automatically applies OCR, generating text content rather than keeping the scanned pages as images in the DOCX file.

For my specific use case, I need the output DOCX file to preserve the scanned PDF pages as images without any text recognition or OCR processing. I couldn't find an option in ExportPDFParams or other related classes to disable OCR.

Is there a way to configure Adobe PDF Services to bypass OCR when converting scanned PDFs to DOCX?

      public bool TryConvertWithAdobe(string filename)
      {
          string docXPath = filename + ".converted.docx";
          var clientId = Environment.GetEnvironmentVariable("ADOBE_SERVICES_CLIENT_ID");
          var secret = Environment.GetEnvironmentVariable("ADOBE_SERVICES_CLIENT_SECRET");
          if (clientId == null || secret == null)
          {
              Logger.Info("Cannot convert pdf to docx, no Adobe credentials found. Skip this step.");
              return false;
          }

          try
          {

              var credentials = new ServicePrincipalCredentials(clientId, secret);
              PDFServices pdfServices = new PDFServices(credentials);

              using Stream inputStream = File.OpenRead(filename);
              IAsset asset = pdfServices.Upload(inputStream, PDFServicesMediaType.PDF.GetMIMETypeValue());

              var exportPDFParams = ExportPDFParams.ExportPDFParamsBuilder(ExportPDFTargetFormat.DOCX)
                .Build();
              var exportPDFJob = new ExportPDFJob(asset, exportPDFParams);

              var location = pdfServices.Submit(exportPDFJob);
              PDFServicesResponse<ExportPDFResult> pdfServicesResponse =
                  pdfServices.GetJobResult<ExportPDFResult>(location, typeof(ExportPDFResult));

              IAsset resultAsset = pdfServicesResponse.Result.Asset;
              StreamAsset streamAsset = pdfServices.GetContent(resultAsset);

              Stream outputStream = File.OpenWrite(docXPath);
              streamAsset.Stream.CopyTo(outputStream);
              outputStream.Close();
              return true;
          }
          catch (Exception ex)
          {
              Logger.Error("Cannot convert pdf to docx with Adobe services: " + ex.Message);
              return false;
          }
      }

Thank you!