Recovering the words on each page
Text recognition —OCR— and AI help recover historical documents, including pages whose initial digitisation produced incomplete text.
← ProjectsNational Library of Spain · Pilot
Written heritage, open to new questions.
We turn digitised pages of historical newspapers and journals into content that can be searched, connected and consulted with AI. A pilot that links answers to their original sources.
OCR · IA · MCP
Giving knowledge structure
We organise articles, pages and passages so information can be found through its content and context. Each piece remains linked to its original publication.
Connecting the archive with AI assistants
Using MCP, a connection protocol, assistants can consult the collection and cite retrieved passages. Readers can return to the original page and check the information.
A pilot with real sources
We worked with five issues of La Época and Nuevo Teatro Crítico. A focused sample to explore new ways of accessing documentary heritage, rather than a digitisation of the entire library collection.
OCR + IA + MCP
From page to enquiry.
One source, new ways to read it.
01 · Text recovered with OCR and AI
Mientras el Sr. Ministro de Gracia y Justicia visita Barcelona para inaugurar solemnemente en la espléndida capital de los antiguos Condes un Palacio de Justicia, los secuestradores andaluces renacen y se entretienen en aprisionar…La justicia en España: antes y después.
La Época · 14/04/1887 · p. 1.
02 · Enquiries with sources
The assistant finds the passage and returns its reference. The original image remains available to check the reading.
View the original page ↗Let’s talk
Let’s talk about your projectTell us about your idea.
We help you shape your project. Together, we will define the scope, priorities and next step.