Search...
  • Pricing
Get Started

Start typing to search across models, sites, and blog posts...

Search across centuries of documents

Once your documents are transcribed, every word becomes findable. Search across thousands of pages in seconds — find names, places, dates, and phrases across entire collections, whether you're working in Transkribus or publishing a searchable website with Sites.

Get started freeExplore published Sites
InstantSearch across all collections
100+Institutions publishing Sites
200M+Pages processed
Full-TextSearch across all transcriptions
Smart SearchFuzzy matching for historical spelling
Tag SearchFilter by named entities and annotations
Public AccessPublish searchable collections with Sites

Full-text search in action

Every transcribed word is indexed. Search across all collections, all documents, all pages — results appear with the exact page and highlighted matches.

Maria Huber 1847
Parish Register St. Jacob, Vol. III, p. 42

den 14. Identity dieses Monaths wurde Maria Huber, Tochter des Johann Huber und der Anna geb. Steiner in der Pfarrkirche zu St. Jacob getauft.

Census Record Innsbruck 1847, p. 118

Maria Huber, 23 Jahre, Dienstmagd, wohnhaft bei Familie Mair, Innstraße 14. Geburtsort: St. Jacob im Walde.

Marriage Register Schwaz, Vol. VII, p. 203

Am 3. September 1847 wurden getraut: Josef Gruber, Sohn des Michael Gruber, und Maria Huber, Tochter des Johann Huber zu St. Jacob.

Correspondence Archive, Letter no. 1847-031

Wir bestätigen den Empfang des Schreibens von Maria Huber betreffend die Erbangelegenheit des verstorbenen Johann Huber.

Try it in Transkribus

Smart Search

Find every spelling variant automatically

Historical handwriting is inconsistent. The same name can appear as 'Müller', 'Mueller', 'Muller', 'Müler', or 'Mühler' within a single collection. Smart Search uses fuzzy matching to find all variants — so you never miss an occurrence. Essential for genealogy, prosopography, and any name-based research.
Fuzzy matching catches spelling variations, OCR errors, and abbreviations
Adjustable similarity threshold — from strict to broad matching
Works across all languages and scripts
Especially powerful for historical name searches
No setup required — built into every Transkribus workspace

Tag Search

Search and filter by named entities

When your documents are annotated with named entities — persons, places, dates, organisations — you unlock a new dimension of search. Filter by entity type, find all mentions of a specific person across an entire archive, or list every place name in a set of letters. Combine tag search with full-text search for precise, structured queries.
Filter by entity type: persons, places, dates, organizations
Cross-reference entities across documents and collections
Combine with full-text search for complex queries
Export entity lists for network analysis and visualization
Manual and AI-powered tagging workflows

Publish searchable collections with Sites

Transkribus Sites turns your transcribed collections into searchable public websites. Visitors can search, browse, and explore your archive — no login required.

habsburger-akten.transkribus.eu
Habsburg Court Archives
Search across all documents…
Document
Transcription
Wir, Ferdinand, von Gottes Gnaden
Erzherzog zu Oesterreich, Herzog
zu Burgundt, Steyr, Kärnten, Crain
und Württemberg, Grave zu Tyrol
entbieten allen und jeden unsern
Underthanen, Ambtleuten und
getreuen Lieben, unser Gnad
und alles Guts. Wir haben
in Erwegung der grossen Noth
dieses Landes beschlossen…

Transkribus Sites

Your archive, searchable by the world

Over 100 institutions already publish their collections as Transkribus Sites. Visitors can search across your entire archive, view original scans side by side with the transcription, browse by collection or tag, and download pages. No web development needed — publish directly from your Transkribus workspace.
Full-text search for visitors — no login required
Side-by-side view: original scan + transcription
Custom domain and branding for your institution
Already used by national archives, university libraries, and museums
Free tier available for small collections
Explore published Sites

From scan to searchable in minutes

The path from a pile of unsearchable scans to a fully indexed, searchable archive is shorter than you think. Upload your documents, run AI text recognition, and every word becomes searchable. For most collections, the entire process is automated — no manual transcription needed.
Upload scans in JPEG, PNG, PDF, or TIFF
Run AI text recognition with 300+ available models
All transcribed text is automatically indexed for search
Search works immediately — no extra setup
Publish as a Site when you're ready to share
See how text recognition works

Make your documents searchable

Start with a free account. Upload your scans, run AI recognition, and search across everything.

Get started freeExplore published Sites
100+Institutions publishing Sites
200M+Pages processed
Free50 credits every month