Appearance

Extraction ​

Extraction turns a document into text or structured data. Upload a file — or pick one you've already saved — and Regoxa reads it for you.

TIP

OCR (Optical Character Recognition) is the technology that reads text from a scanned document or image and turns it into text you can edit, search, and copy.

Add your document ​

  1. In the Extractor module, click Import Files.

    Import Files button

  2. A window opens where you can pick a file you've already saved — under My Files, Recent, or Favorites — or click Create or upload to add a new one.

  3. If you click Create or upload, choose where the file should come from:

    • Browse files — pick a file from your computer.
    • A connected app, such as Google Drive, Microsoft OneDrive, Amazon S3, Cloudflare R2, Google Cloud Storage, Azure Blob Storage, or FTP/SFTP.

    Select input method

    Picking a connected app opens its file list — select the files you want and click Upload.

    Import files from Cloudflare R2

    TIP

    These connected apps are called plugins. Set one up first from Plugins & Integrations so it shows up here.

Choose how to process it ​

Select your file, then pick one of two options at the bottom of the window:

  • Extract Only Text — pulls out the raw text, with no structure. Use this for a quick read of any document.

    Extract only text

  • Select Template To Parse — pulls out specific fields using a template, such as Name, Account Number, or Current Balance. Use this for documents you process regularly, like bank statements or invoices.

    Select a template to parse

    TIP

    A template is a saved pattern that tells Regoxa which fields to look for and where to find them on the page. See Template to create one.

    Pick a template from the list, then click Start Extract.

    Start extraction

Review the results ​

  • Extract Only Text shows the raw text next to your document. Click Apply Deep Mode for a more careful, accurate read, or use Copy and Download to grab the text.

    Extracted text

  • Select Template To Parse shows each field it found — like Name, Account Number, and Current Balance. Click a field to correct it, or use the trash icon to remove one you don't need.

    Extracted fields

    TIP

    You can change template and view template in the extraction page itself if required.

Save your results ​

  • Click Save to store the results in your Workspace. Give the file a name and pick a format — XLS, CSV, JSON, or XML.

    Save results to Workspace

  • Click Plugins to send the results straight to a connected app instead. Choose the plugin and format, then click Send.

    Plugins button

    Send results to a plugin

Tips for better results ​

  • Scan or photograph documents clearly — sharper images give more accurate text
  • Use PDF format for documents with structured data, like tables
  • Turn on Deep Mode for documents where accuracy matters most, like invoices or legal paperwork
  • Always check extracted fields before saving
  • Save time on repeat document types by creating a template

Next steps ​

  • Live Capture — extract text by scanning a physical document with your camera instead of uploading a file
  • Template — create a reusable template for structured field extraction

Video demonstration ​

Watch: Extract images and documents