Transkribus
par READ-COOP
Transkribus is a platform for recognising handwritten and printed text in images of historical documents, run by READ-COOP, a European cooperative in Innsbruck owned by universities, archives, libraries and individuals. Archives, libraries, historians and other researchers use it to transcribe, search and publish manuscripts, registers and printed books, through a web app, an API, an on-premises version and a scanning app. The vendor says neural networks read each line character by character, and that it offers more than 300 public AI models, including its own general model Text Titan, plus layout analysis, table and field extraction, entity tagging and custom model training on as few as 50 transcribed pages. Users correct the recognised text in a built-in editor, and the vendor says recognition accuracy depends mainly on how users train and handle the models.
Score de transparence
Pondéré sur quatre piliers · mis à jour le October 19, 2026
Capacités
Informations sur le fournisseur
Informations complètes sur le fournisseur/prestataire de cette application IA
Infrastructure
Plateformes hébergeant cette application — chacune hérite de sa propre note de transparence.
Informations fournisseur (Règlement IA de l’UE)
Obtenez des informations sur les risques en exécutant des évaluations sur cette application IA.
Vous travaillez chez READ-COOP ? Revendiquez cette fiche pour corriger ou compléter les données.
Prêt à gérer vos applications IA ?
Suivez, évaluez et gouvernez vos applications IA avec Anove.