Security & Data Processing
Updated September 11, 2026. Here is how the current conversion workflow handles your files.
Where image OCR runs
Image OCR is an online service. Your browser sends the image over HTTPS to our server, which submits its contents to Google Cloud Vision for text recognition. The recognized text returns to your browser so you can review, copy or export it. Images larger than 1MB may be compressed before upload.
PDFs and HTML use different workflows
PDF uploads are processed on our server to read their embedded text. Image-only scanned PDFs are not converted directly: save each page as an image and upload that image for OCR. Word and PDF exports are created in your browser from the extracted text.
The HTML-to-PDF converter creates its PDF on your device. We do not send the pasted HTML to our conversion server. Images, fonts or other external URLs in your HTML can still make requests to their own hosts. A conversion-count event does not include your document contents.
Files, results and usage records
The OCR conversion route processes uploaded content in memory and does not write the uploaded file or extracted text to our application database. It does not create a downloadable file history. Save any result you need before leaving the page.
This is separate from account and usage records. We keep conversion counts and timestamps to enforce allowances, using an IP-based identifier for free requests and an account identifier for subscribed requests. Account and subscription information is stored in Firebase, and payments are handled by Paddle. Hosting and processing providers also handle service data under their own terms.
Translation
If you request translation, the selected text is sent to Google Cloud Translation. Translation is a separate processing step with its own usage allowance.
Before uploading sensitive documents
Only upload documents you have permission to process with an online service. Crop unrelated personal information where possible. If your organization requires files to remain entirely on your device, use an approved offline OCR tool.
Questions and account requests
Contact support@extractimagetotext.org with questions about your account or data. Please do not attach sensitive source documents to a support request unless they are necessary and you intend to share them.
See our privacy policy and current plan allowances.