Unlock the past with AI
The operating system for written heritage
built and owned by a community of 250+ stakeholders and 500,000 users.
50 free credits every monthNo credit card required500+ AI models available
Run by a cooperative with 250+ co-owners
Starts in 22 days
Transkribus User Conference 2026
Not all AI is created equal
Generalist AI, or specialised models you can train, inspect and control? Join researchers and archivists who work on documents like yours for three days of talks, workshops and debate.
21–23 September 2026In personOnline
The past wasn't lost, it was just unreadable.
Transkribus turns centuries of handwriting into text you can search, study, and share.
200+ MillionPages deciphered
500.000+Users worldwide
300+Community AI Models
100+Languages Supported
Text Recognition
Upload a scan of any handwritten or printed document. The AI reads it in seconds - across 100+ languages and centuries of scripts.
Handwritten and printed text in 100+ languages
AI-powered, accurate in seconds
Works on centuries-old scripts
Addres to dear Isabella on the Authors
recovery
O Isa pain did visit me
I was at the last extremity
How often did I think of you
I wished your graceful form to view
To clasp you in my weak embrace
Indeed I thought Id run my race
Good Care Im sure was of me taken
But indeed I was much shaken
At last I daily strength did gain
For Archives & Institutions
Make every record searchable
Process massive backlogs of handwritten registers, civil records, and administrative files. AI-powered layout detection and schema-based extraction turn unindexed scans into structured, searchable data.
Full-text search across entire collections
Sovereign EU infrastructure, fully GDPR-compliant
Batch-process millions of pages unattended
For Researchers & DH Projects
From raw scans to citable datasets
Go from manuscript images to structured, publishable data in one collaborative pipeline. Extract entities, annotate your sources, and use Transkribus Sites to turn your results into a public digital edition.
Reproducible workflows for funding compliance
Collaborative annotation with your team
Publish digital editions without coding
For Genealogists & Family Historians
Decipher old handwriting in seconds
Kurrent, French cursive, or faded 19th-century letters - scripts that are nearly impossible to read by hand. Upload a scan, get readable text, and discover your family's story.
300+ community models for every historical script
Works with photos from your phone
No training needed - just upload and read
For Developers & IT Teams
Add text recognition to any workflow
Build on top of Transkribus with a full REST API. Pipe structured outputs directly into your databases, portals, or collection management systems.
REST API with batch processing
Use ready to use models
JSON output ready for any pipeline
Owned by the community, and no one else.
Transkribus is developed by READ-COOP SCE - a European cooperative with over 250+ private and institutional members. Your data stays in Europe, GDPR-compliant, and you retain full ownership.
The past is waiting.
Start analysing historical documents today. 50 free credits every month, no credit card required.