OCR is a multi-stage process. First, the input image is pre-processed: converted to greyscale, binarised (thresholded to black and white), deskewed (rotation corrected), and de-noised. Next, Tesseract's layout analysis segments the image into text regions, lines, and words. Finally, its LSTM model processes each line of text as a sequence of character probabilities and decodes the most likely character sequence using beam search.
Free OCR Tool — Extract Text from Images & PDFs
Optical Character Recognition — Powered by Tesseract
Extract selectable, searchable text from scanned PDFs, photos of documents, and image files. Supports 100+ languages via Tesseract OCR engine.
100+ language support
Scanned PDF & image input
Outputs TXT, DOCX & searchable PDF
Powered by Tesseract OCR (LSTM)
Drop your file here
or browse files
Up to 100 MB free • Outputs: TXT, PDF
Waiting in queue…
Conversion failed
Something went wrong. Please try again.
SSL Encrypted
Files deleted in 2h
Fast processing
No signup needed
Simple & Fast
How to Use OCR Tool
Convert your files in three simple steps — no software, no signup.
01
Upload Image or PDF
Drop a scanned PDF, JPG, PNG, or TIFF of a document. Colour, greyscale, and black-and-white scans all work.
02
Select Language
Choose the language of the text in your document. Multi-language documents can use "Auto-detect" mode.
03
Download Extracted Text
Download as plain TXT, editable DOCX, or searchable PDF with an invisible text layer over the original scan.
Features
Why Use OCR Tool?
100+ Languages
Tesseract supports over 100 languages including Latin, Arabic, Chinese (Simplified/Traditional), Japanese, Korean, and Cyrillic scripts.
Searchable PDF Output
Create a searchable PDF where the original scan is visible but the extracted text layer is selectable and searchable.
DOCX Output
Export recognised text as a Microsoft Word document with basic paragraph structure preserved for further editing.
Image & PDF Input
Accepts JPG, PNG, TIFF, BMP, and multi-page scanned PDF files. Multi-page PDFs process all pages automatically.
Private Processing
Documents are never stored or indexed. OCR runs in a temporary isolated environment and files are deleted within 2 hours.
Pre-processing for Accuracy
Images are automatically deskewed, de-noised, and binarised before OCR to maximise recognition accuracy on real-world scans.
Compatibility
Supported Formats
All the formats you need, all in one place.
| Format | Description | Extension | Best Used For |
|---|---|---|---|
| JPG / JPEG | Photo of a document or scanned page | .jpg |
Smartphone photos of documents |
| PNG | High-contrast screenshot or scan | .png |
Screenshots, receipts, invoices |
| TIFF | Multi-page TIFF scan from scanner | .tiff |
Professional scanner output |
| PDF (SCANNED) | Image-only PDF without text layer | .pdf |
Scanned contracts, books, forms |
| TXT OUTPUT | Plain text extraction from document | .txt |
Data extraction, NLP, databases |
| DOCX OUTPUT | Editable Word document from OCR | .docx |
Edit and format extracted content |
| PDF OUTPUT (OCR) | Searchable PDF with invisible text layer | .pdf |
Archival, full-text search, Ctrl+F |
FAQ
Frequently Asked Questions
OCR (Optical Character Recognition) analyses the pixel patterns in an image to identify characters and words. Tesseract, the engine we use, uses LSTM neural networks for high-accuracy recognition of printed text.
Clean, high-contrast, well-aligned text (like a laser-printed document) achieves 98-99% character accuracy. Handwritten text, low-resolution scans, or unusual fonts will have lower accuracy.
Tesseract can recognise neat, block handwriting but accuracy is much lower than for printed text. For best results use a high-resolution scan (300 dpi+) and select the correct language.
A searchable PDF contains the original scanned image as the visible layer, with an invisible text layer underneath. You can then search (Ctrl+F), select, and copy text in any PDF viewer.
Minimum 200 dpi for acceptable results, 300 dpi recommended for best accuracy. Scans at 600 dpi do not significantly improve accuracy but increase processing time.
Yes. Multi-page PDFs are processed page by page. Each page is de-skewed and recognised individually. Output PDF and DOCX contain all pages in order.
Real-World Uses
Who Uses OCR Tool?
From everyday users to professionals — see how people rely on this tool every day.
Field Worker
Extract Text from Photographed Docs
Photograph a printed form or sign with your phone, upload the image, and get editable text back in seconds.
Archivist
Make Scanned Archives Searchable
Add a searchable text layer to scanned historical documents, contracts, and records for full-text indexing.
Researcher
Extract Text from Academic Papers
OCR scanned journal articles or book chapters to copy passages, run citation searches, and feed into reference managers.
Accountant
Digitise Paper Receipts
Scan expense receipts and OCR them to extract amounts, dates, and merchant names for expense report automation.
Translator
Extract Text for Translation
OCR a scanned document to get editable text, then paste into a translation tool for multilingual document conversion.
Data Engineer
Feed Scanned Data into Pipelines
Convert batches of scanned invoice images to TXT output for automated data extraction and structured data pipelines.
Why HarmonyPal?
HarmonyPal vs. Alternatives
See how we compare to desktop software and other online converters.
| Feature |
Our Tool HarmonyPal |
Adobe Acrobat | Other Online |
|---|---|---|---|
| 100+ language support | Limited | ||
| Searchable PDF output | |||
| DOCX output with structure | |||
| Multi-page PDF input | |||
| Auto deskew & pre-processing | |||
| LSTM neural network engine | Varies | ||
| No watermark on output | |||
| Free to use |
Under the Hood
Technical Specifications
Built on industry-standard open-source tools for maximum quality and reliability.
Limits & Restrictions
- Max file size: 100 MB (free)
- Input: JPG, PNG, TIFF, BMP, PDF (scanned)
- Output: TXT, DOCX, searchable PDF
Tesseract
OCR
Text Recognition
Scanned PDF
LSTM
Conversion Engine
Tesseract 5.x OCR (LSTM neural network) · ImageMagick for pre-processing
Output Quality
98–99% character accuracy on clean 300 dpi+ printed text; lower on handwriting
Average Speed
Typical: 3–8 s per page; multi-page PDFs scale linearly per page
Data Security
HTTPS transfer · isolated temp dir · auto-purged in 2 h
Languages
100+ via Tesseract language packs
Input formats
JPG, PNG, TIFF, BMP, scanned PDF
Output formats
TXT, DOCX, searchable PDF
Complete Guide
How Tesseract OCR Works: LSTM Neural Networks, Pre-Processing, and Language Detection
Tesseract has been developed for over 30 years — originally at HP, then at Google, and now as an open-source project. Its modern LSTM engine brings accuracy levels previously only available in commercial OCR products.
From Image to Text: The OCR Pipeline
LSTM vs Legacy OCR
Older OCR engines (including Tesseract 3.x) used a pattern-matching approach — comparing each character image against a library of stored templates. Tesseract 5's LSTM engine instead trains a recurrent neural network on millions of text examples, learning contextual cues at the line level. This dramatically improves accuracy on cursive text, unusual fonts, and degraded documents where individual characters are ambiguous but their context makes them clear.
Image Pre-Processing for Better Results
Before passing an image to Tesseract, we apply several pre-processing steps via ImageMagick: deskewing corrects documents photographed at an angle; binarisation converts colour or grey images to clean black-and-white text; de-noising removes scanner noise and compression artefacts; contrast enhancement makes faint text darker. These steps can recover several percentage points of accuracy on poor-quality scans.
Searchable PDF Output
A searchable PDF preserves the original scanned image as the visible layer, with an invisible text layer positioned precisely over each word. Generating this correctly requires mapping Tesseract's word bounding boxes (in image pixels) to PDF coordinates (in points, at the original DPI). We use hOCR output from Tesseract — a structured HTML format with bounding box data — to place each word accurately in the output PDF.
Language Support and Multi-Language Documents
Tesseract supports 100+ languages via separate trained data files. Selecting the correct language pack significantly improves accuracy — using an English model on French text will misidentify accented characters. For documents with mixed languages (common in legal documents, academic papers, or multilingual forms), select a combined language model (e.g. eng+fra) or use the auto-detect mode which runs a language identification pass before recognition.