Textora
Turn documents into structured text in seconds.
Upload a scan, PDF, or photo. Textora extracts searchable text you can copy, download, or pipe into your stack.
Drop a file to extract text
PDF, PNG, JPG, WEBP, TIFF — up to your plan limit
Formats & languages
Images and PDFs in English, Persian, Arabic, German, French, Spanish, Dutch, Japanese, Korean, and mixed pairs — with a UI in your language.
Images
JPEG, PNG, WEBP, TIFF, BMP
Documents
Multi-page PDF with per-page text
Exports
TXT, JSON, and plan-gated formats
OCR tools
Purpose-built workspaces for the jobs you run most often.
How it works
Upload, process, review, export — with live job status the whole way.
- 01
Upload
Drop a PDF or image. We validate type and size.
- 02
Process
Jobs move through queued → processing → output.
- 03
Review
Preview the page and edit-ready extracted text.
- 04
Export
Copy, download TXT/JSON, or call the same API.
Encrypted sessions, org-scoped jobs, and clear retention controls so your documents stay yours.
- Cookie sessions with credentialed CORS
- Org-scoped jobs, keys, and webhooks
Developers first
REST endpoints, API keys, and webhooks for production OCR pipelines.
curl -X POST "https://textora.site/api/v1/ocr/jobs" \ -H "Authorization: Bearer $TEXTORA_KEY" \ -F "[email protected]" \ -F "language=eng+fas"Read API docs
Simple plans
Start free. Scale when your volume grows.
Free
Try OCR in the browser
Pro
Higher limits + API access
Business
Batch, webhooks, team seats
Questions
- Which languages are supported?
- English, Persian, Arabic, German, French, Spanish, Dutch, Japanese, Korean, plus mixed pairs such as Eng+Fas and Eng+Ara.
- Do I need an account?
- Yes. OCR tools require a free registered account with a confirmed email. After you sign in, you can upload documents, view history, manage billing, and create API keys.
- Is there an API?
- Yes — REST endpoints with session cookies for the dashboard and API keys for automation.