Florian Langhanki, M.A. -

Reul, Christian ; Nöth, Maximilian ; Baier, Herbert ; Chadbourne, Kevin ; Langhanki, Florian: Human-Centred Open-Source Automatic Text Recognition for the Humanities with OCR4all. In: Humanities-Centred AI (CHAI), 4th Workshop at the 47th German Conference on Artificial Intelligence (2024)
- [ Abstract ]
- [ BibTeX ]
Not least due to the rise of quantitative methods and their need for large volumes of high-quality text, Automatic Text Recognition (ATR) is becoming increasingly important in (Digital) Humanities research. Major technical advances in recent years have been supported by the continuous release of open-source implementations that cover the various individual steps of the ATR workflow. As the optimal approach varies depending on the material at hand and individual quality requirements, the next crucial step is the (further) development of user-friendly software that enables flexible, integrative and sustainable combinations of current and future solutions. To this end, we present the completely revised open-source tool OCR4all, which allows even non-technical users to independently perform an ATR workflow and achieve high-quality results. OCR4all is fully compatible with the ATR solutions from the DFG funding initiative “OCR-D”, so that users have a variety of tools at their disposal that they can use in the best possible way for their specific requirements.

@article{reul2024humancentred, abstract = {Not least due to the rise of quantitative methods and their need for large volumes of high-quality text, Automatic Text Recognition (ATR) is becoming increasingly important in (Digital) Humanities research. Major technical advances in recent years have been supported by the continuous release of open-source implementations that cover the various individual steps of the ATR workflow. As the optimal approach varies depending on the material at hand and individual quality requirements, the next crucial step is the (further) development of user-friendly software that enables flexible, integrative and sustainable combinations of current and future solutions. To this end, we present the completely revised open-source tool OCR4all, which allows even non-technical users to independently perform an ATR workflow and achieve high-quality results. OCR4all is fully compatible with the ATR solutions from the DFG funding initiative “OCR-D”, so that users have a variety of tools at their disposal that they can use in the best possible way for their specific requirements.}, author = {Reul, Christian and Nöth, Maximilian and Baier, Herbert and Chadbourne, Kevin and Langhanki, Florian}, journal = {Humanities-Centred AI (CHAI), 4th Workshop at the 47th German Conference on Artificial Intelligence}, keywords = {myown}, month = {09}, title = {Human-Centred Open-Source Automatic Text Recognition for the Humanities with OCR4all}, year = 2024 }
Langhanki, Florian ; Wehner, Maximilian ; Roeder, Torsten ; Reul, Christian: OCR4all - Open-Source OCR and HTR Across the Centuries., Digital Humanities 2023: Book of Abstracts : Zenodo, 2023
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
OCR4all provides a fully automated OCR workflow by com- bining different OCR solutions. At pretty much any stage of the workflow the user can interact with the results in order to mini- mize consequential errors and optimize the end result. Initially adapted to the difficulties of early modern prints from the 15th and 16th century (project Narragonien digital projects. The soft- ware has undergone continuous development, resulting in signifi- cant advances in layout analysis and highest recognition accuracy through robust OCR models

@proceedings{baillot_anne_2023_7961822, abstract = {OCR4all provides a fully automated OCR workflow by com- bining different OCR solutions. At pretty much any stage of the workflow the user can interact with the results in order to mini- mize consequential errors and optimize the end result. Initially adapted to the difficulties of early modern prints from the 15th and 16th century (project Narragonien digital projects. The soft- ware has undergone continuous development, resulting in signifi- cant advances in layout analysis and highest recognition accuracy through robust OCR models}, author = {Langhanki, Florian and Wehner, Maximilian and Roeder, Torsten and Reul, Christian}, keywords = {myown}, month = {07}, publisher = {Zenodo}, series = {Digital Humanities 2023: Book of Abstracts}, title = {OCR4all - Open-Source OCR and HTR Across the Centuries.}, year = 2023 }
Langhanki, Florian ; Wehner, Maximilian ; Hinrichsen, Lena ; Baierer, Konstantin ; Reul, Christian: Flexibles Arbeiten mit OCR4all - Massenvolltextdigitalisierung von Drucken mithilfe von OCR-D und hochqualitative Transkription von Handschriften. In: DHd2022 Kulturen des digitalen Gedächtnisses (2022)
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
"Die automatisierte Texterkennung von historischen Drucken und Handschriften stellt eine anspruchsvolle Aufgabe bei der Entwicklung modularer wie flexibler OCR-/HTR-Workflows dar. Besonders Forschungsfelder wie Text Mining oder Sentiment Analysis haben die Schwierigkeiten einer Textdigitalisierung entsprechender Materialien bei gleichzeitigem Bedarf großer Textmengen zur Anwendung quantitativer Analyseverfahren erkannt. Auch im Kontext hochqualitativer Volltexterfassungen digitaler Editionen besteht großer Bedarf an intuitiv und komfortabel zu bedienender Software. Die an der Universität Würzburg entwickelte Software OCR4all nimmt neben den formulierten Anforderungen entsprechender Workflowsysteme die Ausrichtung auf einen geisteswissenschaftlichen Nutzer:innenkreis für sich in Anspruch. Durch die baldige Unterstützung der im Rahmen von OCR-D entwickelten Lösungen wird nun die Anwendung im Spannungsfeld einer Massenvolltextdigitalisierung und einer hochqualitativen Erfassung historischer Texte möglich. Der Workshop bietet einen umfassenden Einstieg ins Thema der OCR und HTR historischer Materialien. Teilnehmende werden in die Nutzung von OCR4all eingeführt und dazu befähigt, auf Grundlage anspruchsvoller Ausgangsmaterialien hochqualitative Textdaten zu generieren." Ein Beitrag zur 8. Tagung des Verbands "Digital Humanities im deutschsprachigen Raum" - DHd 2022 Kulturen des digitalen Gedächtnisses.

@article{langhanki_florian_2022_6328079, abstract = {"Die automatisierte Texterkennung von historischen Drucken und Handschriften stellt eine anspruchsvolle Aufgabe bei der Entwicklung modularer wie flexibler OCR-/HTR-Workflows dar. Besonders Forschungsfelder wie Text Mining oder Sentiment Analysis haben die Schwierigkeiten einer Textdigitalisierung entsprechender Materialien bei gleichzeitigem Bedarf großer Textmengen zur Anwendung quantitativer Analyseverfahren erkannt. Auch im Kontext hochqualitativer Volltexterfassungen digitaler Editionen besteht großer Bedarf an intuitiv und komfortabel zu bedienender Software. Die an der Universität Würzburg entwickelte Software OCR4all nimmt neben den formulierten Anforderungen entsprechender Workflowsysteme die Ausrichtung auf einen geisteswissenschaftlichen Nutzer:innenkreis für sich in Anspruch. Durch die baldige Unterstützung der im Rahmen von OCR-D entwickelten Lösungen wird nun die Anwendung im Spannungsfeld einer Massenvolltextdigitalisierung und einer hochqualitativen Erfassung historischer Texte möglich. Der Workshop bietet einen umfassenden Einstieg ins Thema der OCR und HTR historischer Materialien. Teilnehmende werden in die Nutzung von OCR4all eingeführt und dazu befähigt, auf Grundlage anspruchsvoller Ausgangsmaterialien hochqualitative Textdaten zu generieren." Ein Beitrag zur 8. Tagung des Verbands "Digital Humanities im deutschsprachigen Raum" - DHd 2022 Kulturen des digitalen Gedächtnisses.}, author = {Langhanki, Florian and Wehner, Maximilian and Hinrichsen, Lena and Baierer, Konstantin and Reul, Christian}, journal = {DHd2022 Kulturen des digitalen Gedächtnisses}, keywords = {myown}, month = {03}, title = {Flexibles Arbeiten mit OCR4all - Massenvolltextdigitalisierung von Drucken mithilfe von OCR-D und hochqualitative Transkription von Handschriften}, year = 2022 }
Reul, Christian ; Tomasek, Stefan ; Langhanki, Florian ; Springmann, Uwe: Open Source Handwritten Text Recognition on Medieval Manuscripts using Mixed Models and Document-Specific Finetuning. In: arXiv Preprints (submitted to DAS 2022) (2022)
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
This paper deals with the task of practical and open source Handwritten Text Recognition (HTR) on German medieval manuscripts. We report on our efforts to construct mixed recognition models which can be applied out-of-the-box without any further document-specific training but also serve as a starting point for finetuning by training a new model on a few pages of transcribed text (ground truth). To train the mixed models we collected a corpus of 35 manuscripts and ca. 12.5k text lines for two widely used handwriting styles, Gothic and Bastarda cursives. Evaluating the mixed models out-of-the-box on four unseen manuscripts resulted in an average Character Error Rate (CER) of 6.22%. After training on 2, 4 and eventually 32 pages the CER dropped to 3.27%, 2.58%, and 1.65%, respectively. While the in-domain recognition and training of models (Bastarda model to Bastarda material, Gothic to Gothic) unsurprisingly yielded the best results, finetuning out-of-domain models to unseen scripts was still shown to be superior to training from scratch. Our new mixed models have been made openly available to the community.

@article{reul2022open, abstract = {This paper deals with the task of practical and open source Handwritten Text Recognition (HTR) on German medieval manuscripts. We report on our efforts to construct mixed recognition models which can be applied out-of-the-box without any further document-specific training but also serve as a starting point for finetuning by training a new model on a few pages of transcribed text (ground truth). To train the mixed models we collected a corpus of 35 manuscripts and ca. 12.5k text lines for two widely used handwriting styles, Gothic and Bastarda cursives. Evaluating the mixed models out-of-the-box on four unseen manuscripts resulted in an average Character Error Rate (CER) of 6.22%. After training on 2, 4 and eventually 32 pages the CER dropped to 3.27%, 2.58%, and 1.65%, respectively. While the in-domain recognition and training of models (Bastarda model to Bastarda material, Gothic to Gothic) unsurprisingly yielded the best results, finetuning out-of-domain models to unseen scripts was still shown to be superior to training from scratch. Our new mixed models have been made openly available to the community.}, author = {Reul, Christian and Tomasek, Stefan and Langhanki, Florian and Springmann, Uwe}, journal = {arXiv Preprints (submitted to DAS 2022)}, keywords = {myown}, title = {Open Source Handwritten Text Recognition on Medieval Manuscripts using Mixed Models and Document-Specific Finetuning}, year = 2022 }
Jung, Raphaëlle ; Langhanki, Florian ; Wehner, Maximilian: Narragonien digital: ’Den narren spiegel ich diß nenn’. Sebastian Brants Narrenschiff und seine europäische Überlieferungsgeschichte.. In: bildungsgeschichte.de (ed.) , Bibliothek für Bildungsgeschichtliche Forschung (2021)
- [ BibTeX ]
- [ URL ]
@article{https://doi.org/10.25523/32552.9, author = {Jung, Raphaëlle and Langhanki, Florian and Wehner, Maximilian}, editor = {bildungsgeschichte.de}, keywords = {myown}, publisher = {Bibliothek für Bildungsgeschichtliche Forschung}, title = {Narragonien digital: 'Den narren spiegel ich diß nenn'. Sebastian Brants Narrenschiff und seine europäische Überlieferungsgeschichte.}, year = 2021 }
Langhanki, Florian ; Leue, Sebastian ; Hamm, Joachim: ’Das nüv schiff von Narragonia’ (Straßburg [Johannes Grüninger], vor 23.05.1495; GW5048).. In: Burrichter, B. ; Hamm, J. (eds.) Narragonien digital. Digitale Textausgaben von europäischen ’Narrenschiffen’ des 15. Jahrhunderts. (2021)
- [ BibTeX ]
- [ URL ]
@article{leue2021schiff, author = {Langhanki, Florian and Leue, Sebastian and Hamm, Joachim}, editor = {Burrichter, Brigitte and Hamm, Joachim}, journal = {Narragonien digital. Digitale Textausgaben von europäischen 'Narrenschiffen' des 15. Jahrhunderts.}, keywords = {myown}, title = {'Das nüv schiff von Narragonia' (Straßburg [Johannes Grüninger], vor 23.05.1495; GW5048).}, year = 2021 }

11/2024: Game criticism in the 16th century: Das >Spieler-Abc und Namenbüchlein<(Board Games in the Middle Ages - Poetics, Reception, Practice, Fribourg)
02/2024: Basics and Applications of OCR(Studiolo - Digital Humanities Lab 2024; CSMBR Winter Schools: Humanities for the Future)
09/2023: OCR und Edition: Tools und Potenziale(Sprachgeschichte und Editionsphilologie: Vom digitalen Umgang mit historischen Texten, Darmstadt) (with Matthias Boenig)
09/2023: Text recognition with OCR4all (Workshop: AI-supported text recognition for "small" languages; Collegium Carolinum, Munich) (with Niklas Platzer)
12/2021: Code Switching in Sebastian Brant - A series of texts in the Varia Carmina (Transfer and transformation. Dynamics of work and transmission history in the Œuvre of Sebastian Brant. Workshop of the SNF project "Sebastian Brant im Schnittfeld frühneuzeitlicher Textkulturen" (Univ. Bamberg))
05/2021: translator et translatus. Bilingualism in German humanism of the 15th and 16th centuries - The broadsheets and volumes of poetry by Sebastian Brant (Germanist Colloquium of German Philology Würzburg)

02/2024: The right tool for full-text digitisation (DHd24, Passau) (o/b Dr. Christian Reul)
01/2024: Automatic text recognition of prints and manuscripts with OCR4all (HU Berlin)
12/2023: OCR4all - a tool for automatic text recognition of historical sources (Digital Lab, Leipzig) (with Dr Eliese-Sophia Lincke)
11/2023: Open-source text recognition of prints and manuscripts with OCR4all (Digital History Lab, Zurich) (with Dr Christian Reul)
11/2023: Indexing printed and manuscript textual witnesses with OCR4all (Digitale Hermeneutik, WKT Rostock) (with Dr Christian Reul)
11/2023: OCR4all Open-Source Text Recognition of (pre-)modern prints and manuscripts (SUB Hamburg)
11/2023: Open-source text recognition of prints and manuscripts with OCR4all (Praxisatelier Digital Humanities, Zurich) (with Dr Christian Reul)
07/2023: OCR4all - Open-Source OCR and HTR Across the Centuries (DH23 Graz)
05/2023: Practical workshop OCR: Extracting text from images and scans using optical character recognition methods (HAB Wolfenbüttel) (with Anna Lingnau, Lena Hinrichsen, Stefan Büdenbender)
03/2022: Flexible working with OCR4all - Mass full-text digitisation of prints using OCR-D and high-quality transcription of manuscripts (DHd2022, Potsdam) (with Maximilian Wehner, Lena Hinrichsen, Konstantin Baierer)
09/2019: OCR4all - a semi-automatic OCR tool for historical prints (with Maximilian Wehner)
11/2018: OCR on old manuscripts and incunabula (Junior conference "Digital methods and objects in research and mediation of the medieval disciplines" - BaGraMs, Bamberg) (with Dr Michael Dahnke)

2024:

2023:

2022:

2021:

Sebastian Brant and early book printing in Basel. On the 500th anniversary of the death of a humanist scholar.
XVII Old German Studies Colloquium at the Hesselberg
Playful forms of the lyric in the late Middle Ages (1270 - 1500).
Transfer and transformation: Dynamics of work and transmission history in the œuvre of Sebastian Brant.

2018:

Association of Medievalists

AG OCR

Job History

Since 11/2020: Lecturer at the Chair for German Philology of the middle ages and early modern era (University of Würzburg)
Since 10/2022: Research associate at the Centre for Philology and Digitality (University of Würzburg)
10/2021 - 09/2022: Research assistant at the Centre for Philology and Digitality (University of Würzburg)
10/2020 - 10/2021: Research assistant at the Chair for German Philology of the middle ages and early modern era (University of Würzburg)
05/2019 - 12/2019: Student research assistant, project 'Narragonien Digital - Kallimachos' (University of Würzburg)
03/2019 - 09/2020: Student research assistant at the Centre for Philology and Digitality (University of Würzburg)
10/2018 - 07/2020: Tutor at the Chair for German Philology of the middle ages and early modern era (University of Würzburg)
09/2017 - 09/2020: Student research assistant at the Chair for German Philology of the middle ages and early modern era (University of Würzburg)

Education

10/2017 - 10/2020 - Studies of 'Middle Ages and Early Modern Era' (Master) at the University of Würzburg.
- 09/2020 - Master Thesis: "The Trojan War in Print - Development of a Concept for an Online Repository of Early New High German Great Epic"
10/2011 - 02/2018 - Studies of Latin Philology, Greek Philology, German Philology, Geography (Bachelor) at the University of Würzburg.
- 02/2018 - Bachelor Thesis:"Metareference in Herbort von Fritzlar"

Florian Langhanki, M.A.

Further institutional affiliations

Job History

Education

Picture credits