Artificial intelligence
for cultural heritage and archives
AI expertise tailored to each document collection
Since 2014, TEKLIA has been supporting archives, museums, and heritage institutions in their artificial intelligence projects.Whether you manage a cultural institution, a conservation center, or a government agency, our AI solutions adapt to the specific nature of your documents and your business objectives.
They trust us
TEKLIA works with leading heritage institutions, including the Louvre Museum, the French National Archives, the Guimet Museum, the Library of Congress, the Sainte-Geneviève Library and numerous local authorities.
The Impact of TEKLIA’s Technology
Our AI solutions designed for archives, museums and heritage collections analyze documents and media accurately and efficiently, reducing manual work and allowing staff to focus on data interpretation and utilization.
20 million
images processed
33.7 billion
characters transcribed
6x faster
than manual entry
90% savings
compared to manual processing
Discover Your Heritage Collections with Artificial Intelligence
Enhancing heritage collections
Give your collections a new dimension.
Do you possess digitized treasures that are difficult to access?We analyze your images to automatically generate rich metadata and enable full-text search.
✧ Enrich your inventories without tedious manual data entry.
✧ Make your collections searchable by analyzing their content.
Document processing at industrial scale
Scaling up your document processing.
You are familiar with the challenges of HTR and OCR on massive volumes. Teklia equips you with Arkindex, our Data Factory designed to orchestrate the processing of millions of documents. It integrates with your existing systems and guarantees sovereign and reliable document production.
✧ Centralize your workflows: annotations, model comparison, fine-tuning, deployment, quality control, and error recovery in a single tool.
✧ Stay in control: an open-source, flexible platform that can be trained on your own data.
Our Data Factory: The Arkindex Platform
With our Arkindex software, automate large-scale indexing and cataloging. Arkindex is the essential platform for comparing the performance of different AI models on your corpus.
Data Protection and Transparency
We ensure that personal and sensitive data are handled with the utmost care within a secure framework. Arkindex is hosted in Europe and complies with GDPR standards.
Open Source
We publish the source code of our software and our research to facilitate the adoption of AI in the archival and heritage fields.
Responsible AI
Whenever possible, we use smaller and more efficient models to reduce our carbon footprint.
From Raw Images to Actionable Knowledge: Our Processing Workflow
1
Centralize Your Sources
We collect your documents, regardless of their format or location.
______________________
✦ Arkindex organizes your corpora into coherent projects, ready for industrial processing
➙ Flexible bulk import of PDFs and images (JP2, TIFF, JPEG) from your servers, hard drives, or via the IIIF protocol.
2
Reveal Content Through AI
Our algorithms read everything: printed text, ancient manuscripts, and even page structure.
______________________
✦ Arkindex integrates generic models or trains custom models (fine-tuning).
➙ Application of high-performance deep learning models for layout segmentation, HTR, and OCR.
3
Enrich the Data
The software does more than read, it understands. It identifies names, places, and dates to classify your documents.
______________________
✦ Arkindex interprets document content to create metadata.
➙ Transformation of raw text into structured data using Named Entity Recognition (NER) and automatic classification.
4
Enhance and Share
Your collections become searchable via a search engine or ready for online publication.
______________________
✦ Arkindex enables long-term archiving, integration with your databases, or publication on your web portals.
➙ Standardized export (XML-ALTO, METS, TEI, JSON, CSV) or direct integration via our open API.
Start Your Heritage AI Project
Historical and cultural documents require processing aligned with your professional needs.
TEKLIA places process traceability, model performance, and data security at the heart of its AI developments.
Unlock the full potential of your digitized collections!
"We are fully satisfied with our collaboration with the TEKLIA teams. The use of AI allowed us to process a considerable amount of data, it would have been unthinkable for us to rely on manual processing. TEKLIA has always shown itself to be very available and inclined to make the working method evolve, according to the difficulties and problems that arose."
"We are fully satisfied with our collaboration with the TEKLIA teams. The use of AI allowed us to process a considerable amount of data, it would have been unthinkable for us to rely on manual processing. TEKLIA has always shown itself to be very available and inclined to make the working method evolve, according to the difficulties and problems that arose."
"We are very satisfied with our partnership with TEKLIA on the HikarIA project. The team has always been remarkably responsive and flexible. Such agility is invaluable in the development of innovative tools for the automated analysis of photographic collections. Our partnership with TEKLIA will yield a rich new dataset surrounding the Guimet's collections of early photographs of Japan and ensure maximum accessibility."
Latest news
Frequently Asked Questions
Large-scale processing of heritage documents using artificial intelligence takes place after a preliminary digitization phase. TEKLIA supports institutions in implementing reliable industrial solutions for exploiting large heritage corpora, while ensuring rigorous quality control throughout production.
Absolutely.We practice "Fine-tuning": we train our models on your own data to guarantee a faithful transcription of the graphic particularities of your backgrounds.
Yes. Our Arkindex platform provides a comprehensive REST API and supports the IIIF standard, enabling full interoperability with your collection management tools and web portals.