In adeguamento alle norme internazionali la Rivista ha fatto proprio il sistema di accettazione d... more In adeguamento alle norme internazionali la Rivista ha fatto proprio il sistema di accettazione dei saggi attraverso il ricorso sistematico ai referee. I referee rimangono rigorosamente anonimi e sono scelti dalla Fondazione CISAM tra gli studiosi italiani e stranieri maggiormente competenti per i soggetti specifici degli articoli da esaminare. Manoscritti e libri per recensione alla Direzione-Redazione: Studi Medievali, palazzo Racani Arroni, via dell'Arringo -06049 Spoleto (Pg). studimedievali@cisam.org Abbonamenti e vendite alla Fondazione Centro italiano di studi sull'alto medioevo, palazzo Racani Arroni, via dell'Arringo -06049 Spoleto (PG)
Anonimato e pseudoepigrafia nella tradizione latina tardoantica e medievale. Studi in onore di Ileana Pagani, a cura di Francesca Artemisio e Vera Fravventura, Firenze, SISMEL-Edizioni del Galluzzo 2023 (OPA. Opere perdute e anonime [Secoli III-XV], 06), 2023
Lo pseudo Bonaventura. Studi, edizioni e repertorio dei testi e dei manoscritti, 2024
a cura di Laura Vangone La tradizione pseudoepigrafica di Bonaventura attribuisce al francescano,... more a cura di Laura Vangone La tradizione pseudoepigrafica di Bonaventura attribuisce al francescano, tra i vari testi, un Alphabetum in versi. Si tratta di un breve carmen che, nella tradizione manoscritta, appare strettamente legato all'Alphabetum in prosa attribuito al devoto Tommaso da Kempis (1380-1471). Il testo metrico è noto coi titoli di Alphabetum religiosorum incipientium-l'unico a leggersi costantemente nella tradizione manoscritta (d'ora in avanti, ARI)-, Alphabetum religiosorum (metrice scriptum) o Alphabetum minus, per la sua brevità rispetto all'omologo in prosa cui è stato sempre avvicinato; quest'ultimo è indicato generalmente come Alphabetum religiosorum (Vias tuas), Alphabetum monachi o Parvum alphabetum monachi in schola Dei (d'ora in avanti, PAM). L'ARI consta di 24 versi i quali, aprendosi ciascuno su una diversa lettera nell'ordine alfabetico, sciorinano una catena di precetti di buona vita religiosa. Esso è contenutisticamente vicino al De imitatione Christi (n. 36) e ad alcuni testi bonaventuriani, primi tra tutti l'Epistola continens XXV memorialia (n. 7) e la Regula novitiorum (D 53), e certamente ad alcune opere legate alla Devotio moderna quali la Vita boni monachi o l'Epitaphium monachorum. L'ARI si colloca, però, in un genere letterario ancora poco esplorato, soprattutto nella sua declinazione latina, e a cui pure si fece sovente ricorso nei secoli del basso Medioevo per veicolare precetti catechistici e per formare i giovani e i novizi attraverso le formule fissate nella struttura del verso. L'interesse nei confronti di questi testi infatti non è mai decollato realmente, complici anche un valore e una qualità considerati spesso scarsi rispetto ad altri scritti ben più rappresentativi e parlanti. Dopo aver tracciato una breve storia del genere, l'ARI sarà studiato come caso di pseudoepigrafia legato al nome di Bonaventura e ne sarà offerta una nuova edizione. GLI «ALPHABETA» La letteratura spirituale ha spesso fatto ricorso alla serie alfabetica per strutturare i suoi contenuti in prosa ma soprattutto in versi. Ma cos'è precisamente
Poésie et politique dans les mondes nordiques et normands médiévaux (IXe-XIIIe siècle). Actes du colloque de Cerisy-la-Salle (29 septembre-3 octobre 2021), 2024
Le pouvoir et les armes Victor Barabino : Les « rois scaldes » : la formation militaire des souve... more Le pouvoir et les armes Victor Barabino : Les « rois scaldes » : la formation militaire des souverains dans la poésie scaldique des X e-XIII e siècles 21 Alban Gautier : Combattre et commander : le roi et ses guerriers dans la poésie vernaculaire du long X e siècle 39
La Passio Maximi et Venerandi (BHL 5848) est un texte hagiographique peut-être écrit à l’époque d... more La Passio Maximi et Venerandi (BHL 5848) est un texte hagiographique peut-être écrit à l’époque de la Normandie ducale. Son hypotexte est identifié comme la Passio des saints Savin et Cyprien (BHL 7447), ce qui permet de proposer une nouvelle édition critique. L’édition améliore l’identification des sources et la datation de la Passio , qu’on suggère de relier peut-être à l’introduction de l’ordre de Fontevraud en Normandie.
of paper 0695 presented at the Digital Humanities Conference 2019 (DH2019), Utrecht , the Netherl... more of paper 0695 presented at the Digital Humanities Conference 2019 (DH2019), Utrecht , the Netherlands 9-12 July, 2019.
Les humanités numériques accompagnent de plus en plus le travail des chercheurs en leur permettan... more Les humanités numériques accompagnent de plus en plus le travail des chercheurs en leur permettant de mieux mener à terme leurs recherches et de mieux en exploiter les résultats. Ces technologies ont ainsi favorisé la naissance de nombreuses bases de données en ligne. Le ReTeHNor ( Repertorium Textuum Hagiographicorum in Normannia conscriptorum ), rassemblant les textes hagiographiques en latin écrits sur le territoire du duché de Normandie pendant la période ducale (911-1204), est donc l'un de ces répertoires numériques. Basé sur le modèle de la BHL , il rassemble des i ches décrivant au niveau historico-littéraire ces textes hagiographiques et permettant d'avoir, grâce à dif érents modes d'interrogation, une vision critique et statistique de l'hagiographie de Normandie : combien de foyers hagiographiques ? Combien d'hagiographes ? Quels modèles de sainteté ? Quelles typologies d'écrits hagiographiques ? Etc.
La production hagiographique ecrite pendant la periode ducale en Normandie (911-1204) n’a jamais ... more La production hagiographique ecrite pendant la periode ducale en Normandie (911-1204) n’a jamais fait l’objet d’une etude exhaustive et rigoureuse. Il etait impossible d’avoir des donnees statistiques precises et certaines sur les caracteristiques, meme simplement quantitatives, de cette production : sa division par siecles, les rapports entre les textes d’auteurs connus et ceux rediges par des anonymes, la repartition entre le nombre des saints antiques ou medievaux, les sous-genres hagiographiques (vitae, passiones, miracula, translationes), les « typologies de saintete » (eveque, abbe, laic, femme, etc.). Cette situation, prejudiciable a la critique des textes, est due tout d’abord a la difficulte du traitement des textes hagiographiques sur le plan de l’histoire de la litterature: trop d’auteurs restent anonymes, beaucoup de textes ne sont pas dates et leur lieu de production demeure incertain. Ainsi, si l’hagiographie est une source essentielle de l’histoire normande, les condi...
Études sur la sainteté et l'hagiographie-Studies on Sanctity and Hagiography HAGIOLOGIA Études su... more Études sur la sainteté et l'hagiographie-Studies on Sanctity and Hagiography HAGIOLOGIA Études sur la sainteté et l'hagiographie-Studies on Sanctity and Hagiography Des saints et des livres Christianisme fl amboyant et manuscrits hagiographiques du Nord à la fi n du Moyen Âge ( e- e siècle) Études réunies par Fernand Peloux F H 17 Des saints et des livres Des saints et des livres Christianisme fl amboyant et manuscrits hagiographiques du Nord à la fi n du Moyen Âge ( e- e siècle) À la fi n du Moyen Âge, la production hagiographique manuscrite se transforme et connaît son dernier âge d'or entre le succès éditorial de la Légende dorée et l'arrivée de l'imprimerie. De nombreux textes anciens sont abrégés pour intégrer de nouvelles collections. Ce phénomène est en partie responsable du relatif désintérêt des historiens à leur égard : à quoi bon s'intéresser à ces abrégés alors qu'il reste tant à découvrir dans les grands légendiers du Moyen Âge central, et qu'on commence à peine à mieux connaître les tout premiers manuscrits conservés ? L'objectif de ce livre est de mieux saisir la fonction sociale du manuscrit hagiographique, à une période, celle du « christianisme fl amboyant », caractérisée par l'accumulation et la multiplication des dévotions. En se focalisant sur les Pays-Bas méridionaux et une large France septentrionale, une région traversée par la devotio moderna et d'intenses dynamiques religieuses, son objectif est aussi de comprendre ensemble les légendiers latins et vernaculaires, en moyen néerlandais comme dans les parlers d'oïl. Il s'agit de saisir les conditions matérielles de la circulation des textes hagiographiques, mais aussi l'usage de ces manuscrits, dans le cadre de la pastorale et des pratiques cultuelles collectives comme dans celui de l'affi rmation de l'individu à la fi n du Moyen Âge.
DATeCH2019: Proceedings of the 3rd International Conference on Digital Access to Textual Cultural Heritage, 2019
This paper aims to present the first stage of the ANR project VELUM (Towards Innovative Ways of V... more This paper aims to present the first stage of the ANR project VELUM (Towards Innovative Ways of Visualising, Exploring and Linking Resources for Medieval Latin) which, by 2022, is intended to compile the largest representative corpus of Medieval Latin texts. The corpus, which is to comprise 150 millions tokens, is expected to provide selected texts from four centuries of Latin written production (from 800 to 1200 AD) from all across Europe. It will also cover a wide gamut of genres from theological texts to historiography, to documents and letters.
In the first stage of the project, that started in the mid-2018, we are selecting the texts to be included in the corpus, basing on the metadata in the electronic database of Medieval Latin texts that is, at the moment, the largest scholarly-driven source of information of this kind available free on the Internet. Once selected, the texts are retrieved from existing collections and digital libraries.
As early tests showed, less than a half of the texts already exist in interoperable formats such as TEI XML, or at least in a form that allows for easy conversion which does not require human intervention. This means that the bulk of the corpus texts has to be acquired from digital images of editions available on-line through OCR and post-processing. For both tasks, there now exists a broad range of efficient tools, and many sophisticated workflows were proposed in literature. However, the presented project is significantly limited when it comes to its resources, since one person is expected to work on controlling the process and improving OCR quality during a single year.
In the presentation we would like, first, to demonstrate the workflow of the project which, at the moment, consists of the 1) image extraction from PDF files, 2) image cleaning, and its subsequent 3) OCR, followed by 4) the batch-correction of the OCR errors, and 5) the removal of the non-Latin text with a simple classifier. The tools we use are all free and open source, an important factor in a project which is low on resources but ambitious in its goals. The PDF extraction and conversion are performed with Linux 'convert' and 'pdfimages' commands. The output TIFFs are cleaned with the "ScanTailor", while the OCR is realised with "Tesseract". To save on time, the entire workflow is automated, with the human analyst verifying the quality of the output and mass-correcting OCR errors with the "Post Correction Tool".
Apart from presenting the project and the workflow, the paper will discuss the challenges we have faced. One of the most problematic issues turned out to be the relatively disparate quality of the image files retrieved from online sources. Another factor that significantly hinders the automatic processing was the quality of text editions.
In adeguamento alle norme internazionali la Rivista ha fatto proprio il sistema di accettazione d... more In adeguamento alle norme internazionali la Rivista ha fatto proprio il sistema di accettazione dei saggi attraverso il ricorso sistematico ai referee. I referee rimangono rigorosamente anonimi e sono scelti dalla Fondazione CISAM tra gli studiosi italiani e stranieri maggiormente competenti per i soggetti specifici degli articoli da esaminare. Manoscritti e libri per recensione alla Direzione-Redazione: Studi Medievali, palazzo Racani Arroni, via dell'Arringo -06049 Spoleto (Pg). studimedievali@cisam.org Abbonamenti e vendite alla Fondazione Centro italiano di studi sull'alto medioevo, palazzo Racani Arroni, via dell'Arringo -06049 Spoleto (PG)
Anonimato e pseudoepigrafia nella tradizione latina tardoantica e medievale. Studi in onore di Ileana Pagani, a cura di Francesca Artemisio e Vera Fravventura, Firenze, SISMEL-Edizioni del Galluzzo 2023 (OPA. Opere perdute e anonime [Secoli III-XV], 06), 2023
Lo pseudo Bonaventura. Studi, edizioni e repertorio dei testi e dei manoscritti, 2024
a cura di Laura Vangone La tradizione pseudoepigrafica di Bonaventura attribuisce al francescano,... more a cura di Laura Vangone La tradizione pseudoepigrafica di Bonaventura attribuisce al francescano, tra i vari testi, un Alphabetum in versi. Si tratta di un breve carmen che, nella tradizione manoscritta, appare strettamente legato all'Alphabetum in prosa attribuito al devoto Tommaso da Kempis (1380-1471). Il testo metrico è noto coi titoli di Alphabetum religiosorum incipientium-l'unico a leggersi costantemente nella tradizione manoscritta (d'ora in avanti, ARI)-, Alphabetum religiosorum (metrice scriptum) o Alphabetum minus, per la sua brevità rispetto all'omologo in prosa cui è stato sempre avvicinato; quest'ultimo è indicato generalmente come Alphabetum religiosorum (Vias tuas), Alphabetum monachi o Parvum alphabetum monachi in schola Dei (d'ora in avanti, PAM). L'ARI consta di 24 versi i quali, aprendosi ciascuno su una diversa lettera nell'ordine alfabetico, sciorinano una catena di precetti di buona vita religiosa. Esso è contenutisticamente vicino al De imitatione Christi (n. 36) e ad alcuni testi bonaventuriani, primi tra tutti l'Epistola continens XXV memorialia (n. 7) e la Regula novitiorum (D 53), e certamente ad alcune opere legate alla Devotio moderna quali la Vita boni monachi o l'Epitaphium monachorum. L'ARI si colloca, però, in un genere letterario ancora poco esplorato, soprattutto nella sua declinazione latina, e a cui pure si fece sovente ricorso nei secoli del basso Medioevo per veicolare precetti catechistici e per formare i giovani e i novizi attraverso le formule fissate nella struttura del verso. L'interesse nei confronti di questi testi infatti non è mai decollato realmente, complici anche un valore e una qualità considerati spesso scarsi rispetto ad altri scritti ben più rappresentativi e parlanti. Dopo aver tracciato una breve storia del genere, l'ARI sarà studiato come caso di pseudoepigrafia legato al nome di Bonaventura e ne sarà offerta una nuova edizione. GLI «ALPHABETA» La letteratura spirituale ha spesso fatto ricorso alla serie alfabetica per strutturare i suoi contenuti in prosa ma soprattutto in versi. Ma cos'è precisamente
Poésie et politique dans les mondes nordiques et normands médiévaux (IXe-XIIIe siècle). Actes du colloque de Cerisy-la-Salle (29 septembre-3 octobre 2021), 2024
Le pouvoir et les armes Victor Barabino : Les « rois scaldes » : la formation militaire des souve... more Le pouvoir et les armes Victor Barabino : Les « rois scaldes » : la formation militaire des souverains dans la poésie scaldique des X e-XIII e siècles 21 Alban Gautier : Combattre et commander : le roi et ses guerriers dans la poésie vernaculaire du long X e siècle 39
La Passio Maximi et Venerandi (BHL 5848) est un texte hagiographique peut-être écrit à l’époque d... more La Passio Maximi et Venerandi (BHL 5848) est un texte hagiographique peut-être écrit à l’époque de la Normandie ducale. Son hypotexte est identifié comme la Passio des saints Savin et Cyprien (BHL 7447), ce qui permet de proposer une nouvelle édition critique. L’édition améliore l’identification des sources et la datation de la Passio , qu’on suggère de relier peut-être à l’introduction de l’ordre de Fontevraud en Normandie.
of paper 0695 presented at the Digital Humanities Conference 2019 (DH2019), Utrecht , the Netherl... more of paper 0695 presented at the Digital Humanities Conference 2019 (DH2019), Utrecht , the Netherlands 9-12 July, 2019.
Les humanités numériques accompagnent de plus en plus le travail des chercheurs en leur permettan... more Les humanités numériques accompagnent de plus en plus le travail des chercheurs en leur permettant de mieux mener à terme leurs recherches et de mieux en exploiter les résultats. Ces technologies ont ainsi favorisé la naissance de nombreuses bases de données en ligne. Le ReTeHNor ( Repertorium Textuum Hagiographicorum in Normannia conscriptorum ), rassemblant les textes hagiographiques en latin écrits sur le territoire du duché de Normandie pendant la période ducale (911-1204), est donc l'un de ces répertoires numériques. Basé sur le modèle de la BHL , il rassemble des i ches décrivant au niveau historico-littéraire ces textes hagiographiques et permettant d'avoir, grâce à dif érents modes d'interrogation, une vision critique et statistique de l'hagiographie de Normandie : combien de foyers hagiographiques ? Combien d'hagiographes ? Quels modèles de sainteté ? Quelles typologies d'écrits hagiographiques ? Etc.
La production hagiographique ecrite pendant la periode ducale en Normandie (911-1204) n’a jamais ... more La production hagiographique ecrite pendant la periode ducale en Normandie (911-1204) n’a jamais fait l’objet d’une etude exhaustive et rigoureuse. Il etait impossible d’avoir des donnees statistiques precises et certaines sur les caracteristiques, meme simplement quantitatives, de cette production : sa division par siecles, les rapports entre les textes d’auteurs connus et ceux rediges par des anonymes, la repartition entre le nombre des saints antiques ou medievaux, les sous-genres hagiographiques (vitae, passiones, miracula, translationes), les « typologies de saintete » (eveque, abbe, laic, femme, etc.). Cette situation, prejudiciable a la critique des textes, est due tout d’abord a la difficulte du traitement des textes hagiographiques sur le plan de l’histoire de la litterature: trop d’auteurs restent anonymes, beaucoup de textes ne sont pas dates et leur lieu de production demeure incertain. Ainsi, si l’hagiographie est une source essentielle de l’histoire normande, les condi...
Études sur la sainteté et l'hagiographie-Studies on Sanctity and Hagiography HAGIOLOGIA Études su... more Études sur la sainteté et l'hagiographie-Studies on Sanctity and Hagiography HAGIOLOGIA Études sur la sainteté et l'hagiographie-Studies on Sanctity and Hagiography Des saints et des livres Christianisme fl amboyant et manuscrits hagiographiques du Nord à la fi n du Moyen Âge ( e- e siècle) Études réunies par Fernand Peloux F H 17 Des saints et des livres Des saints et des livres Christianisme fl amboyant et manuscrits hagiographiques du Nord à la fi n du Moyen Âge ( e- e siècle) À la fi n du Moyen Âge, la production hagiographique manuscrite se transforme et connaît son dernier âge d'or entre le succès éditorial de la Légende dorée et l'arrivée de l'imprimerie. De nombreux textes anciens sont abrégés pour intégrer de nouvelles collections. Ce phénomène est en partie responsable du relatif désintérêt des historiens à leur égard : à quoi bon s'intéresser à ces abrégés alors qu'il reste tant à découvrir dans les grands légendiers du Moyen Âge central, et qu'on commence à peine à mieux connaître les tout premiers manuscrits conservés ? L'objectif de ce livre est de mieux saisir la fonction sociale du manuscrit hagiographique, à une période, celle du « christianisme fl amboyant », caractérisée par l'accumulation et la multiplication des dévotions. En se focalisant sur les Pays-Bas méridionaux et une large France septentrionale, une région traversée par la devotio moderna et d'intenses dynamiques religieuses, son objectif est aussi de comprendre ensemble les légendiers latins et vernaculaires, en moyen néerlandais comme dans les parlers d'oïl. Il s'agit de saisir les conditions matérielles de la circulation des textes hagiographiques, mais aussi l'usage de ces manuscrits, dans le cadre de la pastorale et des pratiques cultuelles collectives comme dans celui de l'affi rmation de l'individu à la fi n du Moyen Âge.
DATeCH2019: Proceedings of the 3rd International Conference on Digital Access to Textual Cultural Heritage, 2019
This paper aims to present the first stage of the ANR project VELUM (Towards Innovative Ways of V... more This paper aims to present the first stage of the ANR project VELUM (Towards Innovative Ways of Visualising, Exploring and Linking Resources for Medieval Latin) which, by 2022, is intended to compile the largest representative corpus of Medieval Latin texts. The corpus, which is to comprise 150 millions tokens, is expected to provide selected texts from four centuries of Latin written production (from 800 to 1200 AD) from all across Europe. It will also cover a wide gamut of genres from theological texts to historiography, to documents and letters.
In the first stage of the project, that started in the mid-2018, we are selecting the texts to be included in the corpus, basing on the metadata in the electronic database of Medieval Latin texts that is, at the moment, the largest scholarly-driven source of information of this kind available free on the Internet. Once selected, the texts are retrieved from existing collections and digital libraries.
As early tests showed, less than a half of the texts already exist in interoperable formats such as TEI XML, or at least in a form that allows for easy conversion which does not require human intervention. This means that the bulk of the corpus texts has to be acquired from digital images of editions available on-line through OCR and post-processing. For both tasks, there now exists a broad range of efficient tools, and many sophisticated workflows were proposed in literature. However, the presented project is significantly limited when it comes to its resources, since one person is expected to work on controlling the process and improving OCR quality during a single year.
In the presentation we would like, first, to demonstrate the workflow of the project which, at the moment, consists of the 1) image extraction from PDF files, 2) image cleaning, and its subsequent 3) OCR, followed by 4) the batch-correction of the OCR errors, and 5) the removal of the non-Latin text with a simple classifier. The tools we use are all free and open source, an important factor in a project which is low on resources but ambitious in its goals. The PDF extraction and conversion are performed with Linux 'convert' and 'pdfimages' commands. The output TIFFs are cleaned with the "ScanTailor", while the OCR is realised with "Tesseract". To save on time, the entire workflow is automated, with the human analyst verifying the quality of the output and mass-correcting OCR errors with the "Post Correction Tool".
Apart from presenting the project and the workflow, the paper will discuss the challenges we have faced. One of the most problematic issues turned out to be the relatively disparate quality of the image files retrieved from online sources. Another factor that significantly hinders the automatic processing was the quality of text editions.
XXIV CORSO INTERNAZIONALE DI FORMAZIONE BIBLIOGRAFICA
Medioevo latino. Metodologie e tecniche bib... more XXIV CORSO INTERNAZIONALE DI FORMAZIONE BIBLIOGRAFICA Medioevo latino. Metodologie e tecniche bibliografiche Firenze, 1 ottobre 2020
The ANR-funded project (2018-2022) is a first step towards an innovative digital environment for ... more The ANR-funded project (2018-2022) is a first step towards an innovative digital environment for the study of the language and culture of medieval Europe. It has a challenging and ambitious goal: to build foundations for empirical research in medieval culture, language and history, by providing an appropriate environment for textual source analysis. We aim at building, firstly, a large (100 M tokens) balanced corpus of Medieval Latin texts composed between 500 and 1500 AD all across Europe. In order to assess the population that we would like our corpus to reflect, we rely mainly on the most comprehensive list of Medieval Latin texts, namely the Index Scriptorum Mediae Latinitatis (Blatt, 1973; Bon, 2005).
Uploads
Papers by Laura Vangone
In the first stage of the project, that started in the mid-2018, we are selecting the texts to be included in the corpus, basing on the metadata in the electronic database of Medieval Latin texts that is, at the moment, the largest scholarly-driven source of information of this kind available free on the Internet. Once selected, the texts are retrieved from existing collections and digital libraries.
As early tests showed, less than a half of the texts already exist in interoperable formats such as TEI XML, or at least in a form that allows for easy conversion which does not require human intervention. This means that the bulk of the corpus texts has to be acquired from digital images of editions available on-line through OCR and post-processing. For both tasks, there now exists a broad range of efficient tools, and many sophisticated workflows were proposed in literature. However, the presented project is significantly limited when it comes to its resources, since one person is expected to work on controlling the process and improving OCR quality during a single year.
In the presentation we would like, first, to demonstrate the workflow of the project which, at the moment, consists of the 1) image extraction from PDF files, 2) image cleaning, and its subsequent 3) OCR, followed by 4) the batch-correction of the OCR errors, and 5) the removal of the non-Latin text with a simple classifier. The tools we use are all free and open source, an important factor in a project which is low on resources but ambitious in its goals. The PDF extraction and conversion are performed with Linux 'convert' and 'pdfimages' commands. The output TIFFs are cleaned with the "ScanTailor", while the OCR is realised with "Tesseract". To save on time, the entire workflow is automated, with the human analyst verifying the quality of the output and mass-correcting OCR errors with the "Post Correction Tool".
Apart from presenting the project and the workflow, the paper will discuss the challenges we have faced. One of the most problematic issues turned out to be the relatively disparate quality of the image files retrieved from online sources. Another factor that significantly hinders the automatic processing was the quality of text editions.
In the first stage of the project, that started in the mid-2018, we are selecting the texts to be included in the corpus, basing on the metadata in the electronic database of Medieval Latin texts that is, at the moment, the largest scholarly-driven source of information of this kind available free on the Internet. Once selected, the texts are retrieved from existing collections and digital libraries.
As early tests showed, less than a half of the texts already exist in interoperable formats such as TEI XML, or at least in a form that allows for easy conversion which does not require human intervention. This means that the bulk of the corpus texts has to be acquired from digital images of editions available on-line through OCR and post-processing. For both tasks, there now exists a broad range of efficient tools, and many sophisticated workflows were proposed in literature. However, the presented project is significantly limited when it comes to its resources, since one person is expected to work on controlling the process and improving OCR quality during a single year.
In the presentation we would like, first, to demonstrate the workflow of the project which, at the moment, consists of the 1) image extraction from PDF files, 2) image cleaning, and its subsequent 3) OCR, followed by 4) the batch-correction of the OCR errors, and 5) the removal of the non-Latin text with a simple classifier. The tools we use are all free and open source, an important factor in a project which is low on resources but ambitious in its goals. The PDF extraction and conversion are performed with Linux 'convert' and 'pdfimages' commands. The output TIFFs are cleaned with the "ScanTailor", while the OCR is realised with "Tesseract". To save on time, the entire workflow is automated, with the human analyst verifying the quality of the output and mass-correcting OCR errors with the "Post Correction Tool".
Apart from presenting the project and the workflow, the paper will discuss the challenges we have faced. One of the most problematic issues turned out to be the relatively disparate quality of the image files retrieved from online sources. Another factor that significantly hinders the automatic processing was the quality of text editions.
Medioevo latino. Metodologie e tecniche bibliografiche
Firenze, 1 ottobre 2020