Automated Handwritten Text Recognition of Manuscript Document Collections скачать в хорошем качестве

Automated Handwritten Text Recognition of Manuscript Document Collections

Digitized handwritten manuscripts often remain hidden because they lack searchable text, and manual transcription is prohibitively time-consuming and unsustainable at scale. Generative artificial tools (AI) now offer an opportunity to automate this process and greatly expand access to these valuable resources. Lehigh University Libraries has developed a digital repository service that leverages generative AI models with advanced handwritten text recognition (HTR) capabilities to transcribe handwritten materials. The Libraries received the 2025 Lyrasis Catalyst Grant to advance this work by developing an open-source application that empowers the broader library community to use generative AI for automated transcription and indexing of digitized manuscripts. This project briefing presents the progress to date, highlights the tools and processes used to evaluate different large language models, and outlines next steps toward creating an accurate and cost-effective solution to enable libraries to automatically transcribe and index handwritten materials. Chulin Meng, Lehigh University Joe Corall, Lehigh University CNI Pre-Recorded Project Briefing Series Winter 2026 Edition More information: https://www.cni.org/news/ppbs_edition... *Subscribe to our channel*: https://www.youtube.com/c/cnivideo?su... Stay connected with us Website: https://www.cni.org Facebook:   / cni.org   LinkedIn:   / coalition-for-networked-information   BlueSky: https://bsky.app/profile/cni-org.bsky... Mastodon: https://mastodon.social/@cni X: https://x.com/cni_org Subscribe to our listserv CNI-ANNOUNCE: https://www.cni.org/resources/follow-...