Today at our Winter School on HTR of Historical Sources, we're working with eScriptorium, an open-source platform for transcribing historical documents!
Many thanks to @inriaparisnlp.bsky.social for letting us use their instance and supporting our school! 🙏
#DigitalHumanities #HTR #eScriptorium
Many thanks to @inriaparisnlp.bsky.social for letting us use their instance and supporting our school! 🙏
#DigitalHumanities #HTR #eScriptorium
October 30, 2026 at 12:37 PM
Today at our Winter School on HTR of Historical Sources, we're working with eScriptorium, an open-source platform for transcribing historical documents!
Many thanks to @inriaparisnlp.bsky.social for letting us use their instance and supporting our school! 🙏
#DigitalHumanities #HTR #eScriptorium
Many thanks to @inriaparisnlp.bsky.social for letting us use their instance and supporting our school! 🙏
#DigitalHumanities #HTR #eScriptorium
The app also keeps the PAGE XML format 💪 Because then the interpretative text stays line by line, linked to the regions and coordinates of the original diplomatic HTR. You can open it in Transkribus or eScriptorium and check every reading against the manuscript, not just a loose plain-text edition.
November 4, 2026 at 3:16 PM
The app also keeps the PAGE XML format 💪 Because then the interpretative text stays line by line, linked to the regions and coordinates of the original diplomatic HTR. You can open it in Transkribus or eScriptorium and check every reading against the manuscript, not just a loose plain-text edition.
New release of OICEN-HTR #OldIcelandic #HTR is now available in three flavours:
1) CATMuS zenodo.org/records/2301...
2) PP-OCRv6-small zenodo.org/records/2303...
3) PP-OCRv6-medium zenodo.org/records/2321...
1560 images, 56438 lines, 713267 words, and 2358651 characters from 15 mss. and ~35 fragm.
1) CATMuS zenodo.org/records/2301...
2) PP-OCRv6-small zenodo.org/records/2303...
3) PP-OCRv6-medium zenodo.org/records/2321...
1560 images, 56438 lines, 713267 words, and 2358651 characters from 15 mss. and ~35 fragm.
October 28, 2026 at 1:12 PM
New release of OICEN-HTR #OldIcelandic #HTR is now available in three flavours:
1) CATMuS zenodo.org/records/2301...
2) PP-OCRv6-small zenodo.org/records/2303...
3) PP-OCRv6-medium zenodo.org/records/2321...
1560 images, 56438 lines, 713267 words, and 2358651 characters from 15 mss. and ~35 fragm.
1) CATMuS zenodo.org/records/2301...
2) PP-OCRv6-small zenodo.org/records/2303...
3) PP-OCRv6-medium zenodo.org/records/2321...
1560 images, 56438 lines, 713267 words, and 2358651 characters from 15 mss. and ~35 fragm.
✍🏻 Va ser un veritable luxe passar una estona de dimarts al vespre amb els alumnes del col·lectiu Escriptorium de Sabadell per parlar plegats de contes, de referents i d’escriptura, i disseccionar plegats «Animals inexpressius» (@laltraeditorial.bsky.social).
October 16, 2026 at 11:25 AM
✍🏻 Va ser un veritable luxe passar una estona de dimarts al vespre amb els alumnes del col·lectiu Escriptorium de Sabadell per parlar plegats de contes, de referents i d’escriptura, i disseccionar plegats «Animals inexpressius» (@laltraeditorial.bsky.social).
The other corpus I work on, the Jerusalem court registers, the sijillāt, has no machine-readable text at all. There the convention is not inherited; we have to set it ourselves. That turns out to be the founding act of the whole project rather than a technical detail.
October 9, 2026 at 2:09 PM
The other corpus I work on, the Jerusalem court registers, the sijillāt, has no machine-readable text at all. There the convention is not inherited; we have to set it ourselves. That turns out to be the founding act of the whole project rather than a technical detail.
1. eScriptorium 🆓
I use it daily. It is open-source, browser-based, and strong on right-to-left and non-Latin scripts. There is a catch: you need a server to run it, and shared instances often ask you to release your data and models openly in return.
I use it daily. It is open-source, browser-based, and strong on right-to-left and non-Latin scripts. There is a catch: you need a server to run it, and shared instances often ask you to release your data and models openly in return.
October 6, 2026 at 11:19 AM
1. eScriptorium 🆓
I use it daily. It is open-source, browser-based, and strong on right-to-left and non-Latin scripts. There is a catch: you need a server to run it, and shared instances often ask you to release your data and models openly in return.
I use it daily. It is open-source, browser-based, and strong on right-to-left and non-Latin scripts. There is a catch: you need a server to run it, and shared instances often ask you to release your data and models openly in return.
There is no single best. My own preference is eScriptorium. Personally, the technique is not that hard to acquire. For a beginner, though, I'd point to Transkribus: it is easier to start with, and its cooperative model sits better with me than a purely commercial one.
October 6, 2026 at 11:19 AM
There is no single best. My own preference is eScriptorium. Personally, the technique is not that hard to acquire. For a beginner, though, I'd point to Transkribus: it is easier to start with, and its cooperative model sits better with me than a purely commercial one.
[PUBLICATION : ] Dans le cadre du projet TMD (Tours municipal Data), publication de modèles HTR pour la transcription d'écritures cursives des XVe et XVIe siècles.
Travail produit sous eScriptorium par D. Nivarlet et Cl. Leroy.
renumar.univ-tours.fr/publication/...
Travail produit sous eScriptorium par D. Nivarlet et Cl. Leroy.
renumar.univ-tours.fr/publication/...
Modèles de transcription HTR pour écritures des XVe et XVIe siècles – renumar
renumar.univ-tours.fr
September 15, 2026 at 4:57 PM
[PUBLICATION : ] Dans le cadre du projet TMD (Tours municipal Data), publication de modèles HTR pour la transcription d'écritures cursives des XVe et XVIe siècles.
Travail produit sous eScriptorium par D. Nivarlet et Cl. Leroy.
renumar.univ-tours.fr/publication/...
Travail produit sous eScriptorium par D. Nivarlet et Cl. Leroy.
renumar.univ-tours.fr/publication/...
#eScriptorium 26.07 is out since yesterday, and our maintainer just posted the blog post about it : escriptorium.eu/blog/2026-09...
And there is something we (the eScriptorium Steering Committee) wanted to talk to you about.
And there is something we (the eScriptorium Steering Committee) wanted to talk to you about.
July 2026 Release
The 26.07 release brings document-scoped ontologies, a new Overview panel for regions and lines, an Ontology Overview page for quality control, custom transcription fonts, full document archives and a...
escriptorium.eu
September 4, 2026 at 8:11 AM
#eScriptorium 26.07 is out since yesterday, and our maintainer just posted the blog post about it : escriptorium.eu/blog/2026-09...
And there is something we (the eScriptorium Steering Committee) wanted to talk to you about.
And there is something we (the eScriptorium Steering Committee) wanted to talk to you about.
2/ For the first time in the history of eScriptorium, and maybe for a project in DH (?), we found security issues that were bad. And I mean very bad for some.
This led us to record 5 CVE (Common Vulnerabilities and Exposures) and learn to follow a protocol we never had to deal with before.
This led us to record 5 CVE (Common Vulnerabilities and Exposures) and learn to follow a protocol we never had to deal with before.
September 4, 2026 at 8:11 AM
2/ For the first time in the history of eScriptorium, and maybe for a project in DH (?), we found security issues that were bad. And I mean very bad for some.
This led us to record 5 CVE (Common Vulnerabilities and Exposures) and learn to follow a protocol we never had to deal with before.
This led us to record 5 CVE (Common Vulnerabilities and Exposures) and learn to follow a protocol we never had to deal with before.
3/ Because we are an open source software, people install it in place we don't control: that's basically the goal of the decentralization behind eScriptorium.
So we tried to reach out and warn all known instances of the patch (and in some case help them).
From now on, CVEs will be recorded.
So we tried to reach out and warn all known instances of the patch (and in some case help them).
From now on, CVEs will be recorded.
September 4, 2026 at 8:11 AM
3/ Because we are an open source software, people install it in place we don't control: that's basically the goal of the decentralization behind eScriptorium.
So we tried to reach out and warn all known instances of the patch (and in some case help them).
From now on, CVEs will be recorded.
So we tried to reach out and warn all known instances of the patch (and in some case help them).
From now on, CVEs will be recorded.
4/ If you didn't see the articles lately, AI is there in both development and security. Linux Kernels went from an average of 500 CVE / release to 1250 and then 2000 in the span of 4 releases.
And let's say we can see that agentic coding has an impact on our work (and we have only one Hassen).
And let's say we can see that agentic coding has an impact on our work (and we have only one Hassen).
September 4, 2026 at 8:11 AM
4/ If you didn't see the articles lately, AI is there in both development and security. Linux Kernels went from an average of 500 CVE / release to 1250 and then 2000 in the span of 4 releases.
And let's say we can see that agentic coding has an impact on our work (and we have only one Hassen).
And let's say we can see that agentic coding has an impact on our work (and we have only one Hassen).
this is either already OCRed stuff or whatever native capacity is in AnythingLLM; most of my stuff is either published material-- a treatise by Jean Paul Marat from 1773-- or manuscript where I need some kind of transcription thing. working on escriptorium for that
September 3, 2026 at 5:09 PM
this is either already OCRed stuff or whatever native capacity is in AnythingLLM; most of my stuff is either published material-- a treatise by Jean Paul Marat from 1773-- or manuscript where I need some kind of transcription thing. working on escriptorium for that
After multiple setbacks during the summer, eScriptorium 26.7 will be out after its month long beta test on Friday.
Stay tuned
Stay tuned
September 2, 2026 at 4:12 PM
After multiple setbacks during the summer, eScriptorium 26.7 will be out after its month long beta test on Friday.
Stay tuned
Stay tuned
Los colegas de #escriptorium , el mismo software que vamos a usar nosotros, cc/ @ponteineptique.bsky.social
Un buen ejemplo tanto de uso útil de IA en Historia, como de su fiabilidad cuestionable y la imposibilidad de prescindir del especialista humano. Abre mucho campo, aunque también lo abre a la construcción de discurso a partir de errores no comprobados. Tasas de hasta el 20% de error dan que pensar.
September 2, 2026 at 5:08 PM
Los colegas de #escriptorium , el mismo software que vamos a usar nosotros, cc/ @ponteineptique.bsky.social
eScriptorium continue de buguer à mort sur l'ordi fixe ceci dit, ça me rend chèvre même si c'est beaucoup mieux qu'en mai-juin (et qu'on m'a visiblement filé un modèle de segmentation cool). J'ai fait un retour, pas eu de réponse, après je soupçonne que ça soit une toute petite équipe sur le projet.
August 19, 2026 at 9:27 AM
eScriptorium continue de buguer à mort sur l'ordi fixe ceci dit, ça me rend chèvre même si c'est beaucoup mieux qu'en mai-juin (et qu'on m'a visiblement filé un modèle de segmentation cool). J'ai fait un retour, pas eu de réponse, après je soupçonne que ça soit une toute petite équipe sur le projet.
#Manuscript Studies (SIMS)#Manuscript Cultures (CSMC)#eScriptorium platform to build machine-learning models that can transcribe handwritten manuscripts
Reading manuscripts in the digital space: How the Penn Libraries is advancing Handwritten Text Recognition
Staff and fellows are using the platform eScriptorium to build machine-learning models that can transcribe handwritten manuscripts from across the world.
www.library.upenn.edu
August 18, 2026 at 11:08 PM
#Manuscript Studies (SIMS)#Manuscript Cultures (CSMC)#eScriptorium platform to build machine-learning models that can transcribe handwritten manuscripts
Watch the talks from spring 2026 here: youtube.com/playlist?lis...
August 12, 2026 at 3:35 AM
Watch the talks from spring 2026 here: youtube.com/playlist?lis...
Crispons nous sur eScriptorium depuis 4/5 jours.
July 18, 2025 at 11:03 AM
Crispons nous sur eScriptorium depuis 4/5 jours.
First workshop of the #escriptorium server installed at the @uva-es.bsky.social. Having some issues with #CATMuS and this cursive calligraphy.
March 14, 2025 at 9:27 AM
First workshop of the #escriptorium server installed at the @uva-es.bsky.social. Having some issues with #CATMuS and this cursive calligraphy.
La Universidad de Valladolid, pionera en el uso de IA para la transcripción de manuscritos.
La Universidad de Valladolid (UVA), primera de España en utilizar la plataforma ‘eScriptorum’ de transcripción digital de textos con IA.
Detalle del 'Libro de los judizios' UVA La Universidad de Valladolid (UVa) pone en marcha la innovadora plataforma de inteligencia artificial ‘eScriptorium’ diseñada para la transcripción automatizada de libros antiguos y de todo tipo de manuscritos digitalizados. Se trata de un nodo con carácter experimental que permitirá fortalecer la conservación y recuperación de archivos documentales y supone un avance para la investigación en Humanidades Digitales.
bibliotecahistoricausal.wordpress.com
March 27, 2025 at 10:55 AM
La Universidad de Valladolid, pionera en el uso de IA para la transcripción de manuscritos.
Im nächsten DigitalHistoryOFK diskutiert Pauline Spychala (DHI Paris) wie eScriptorium & @transkribus.bsky.social effektiv kombiniert werden können, um mittelalterliche franz. Quellen per HTR zu erschließen.
📆Mi, 22. Nov., 16-18 Uhr, Zoom (offen für alle!)
🔗Abstract: dhistory.hypotheses.org/6384
📆Mi, 22. Nov., 16-18 Uhr, Zoom (offen für alle!)
🔗Abstract: dhistory.hypotheses.org/6384
November 21, 2023 at 5:34 PM
Im nächsten DigitalHistoryOFK diskutiert Pauline Spychala (DHI Paris) wie eScriptorium & @transkribus.bsky.social effektiv kombiniert werden können, um mittelalterliche franz. Quellen per HTR zu erschließen.
📆Mi, 22. Nov., 16-18 Uhr, Zoom (offen für alle!)
🔗Abstract: dhistory.hypotheses.org/6384
📆Mi, 22. Nov., 16-18 Uhr, Zoom (offen für alle!)
🔗Abstract: dhistory.hypotheses.org/6384
🟠 CVE-2026-18359 - High (8.5)
Server-side request forgery in the METS and IIIF import URI handling in Scripta eScriptorium thro...
https://www.thehackerwire.com/vulnerability/CVE-2026-18359/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
Server-side request forgery in the METS and IIIF import URI handling in Scripta eScriptorium thro...
https://www.thehackerwire.com/vulnerability/CVE-2026-18359/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
August 6, 2026 at 7:00 PM
🟠 CVE-2026-18359 - High (8.5)
Server-side request forgery in the METS and IIIF import URI handling in Scripta eScriptorium thro...
https://www.thehackerwire.com/vulnerability/CVE-2026-18359/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
Server-side request forgery in the METS and IIIF import URI handling in Scripta eScriptorium thro...
https://www.thehackerwire.com/vulnerability/CVE-2026-18359/
#infosec #cybersecurity #CVE #vulnerability #security #patchstack
Bookmarked: eScriptorium https://escriptorium.eu/blog/2026-04-15-a-new-chapter-for-escriptorium/ #dh #htr #handschriften A Digital Text Production Pipeline for Print and Handwritten Texts using machine learning techniques.
A New Chapter for eScriptorium
In the next few months you will likely be seeing a new version of eScriptorium come to your instance. If you’re using INRIA, you may have already seen the changes. This year has brought a new milestone with major additions and improvements to eScriptorium. These updates were so significant that we had even started naming this release ‘version 1.0’, (and they are even so significant that starting version 1.0, we’ll change our naming scheme, see below).
escriptorium.eu
April 22, 2026 at 7:35 PM
Bookmarked: eScriptorium https://escriptorium.eu/blog/2026-04-15-a-new-chapter-for-escriptorium/ #dh #htr #handschriften A Digital Text Production Pipeline for Print and Handwritten Texts using machine learning techniques.
Bookmarked: Online Tools for Handwritten Text Recognition: A Comparative Review of Transkribus and eScriptorium for Byzantine Paleography - The Stoa: a Review for Digital ClassicsThe Stoa: a Review for Digital Classics https://blog.stoa.org/archives/4308 #htr_transkribus_escriptorium This […]
Original post on fedihum.org
fedihum.org
February 2, 2025 at 10:37 AM
Bookmarked: Online Tools for Handwritten Text Recognition: A Comparative Review of Transkribus and eScriptorium for Byzantine Paleography - The Stoa: a Review for Digital ClassicsThe Stoa: a Review for Digital Classics https://blog.stoa.org/archives/4308 #htr_transkribus_escriptorium This […]