Getting Started with Transkribus Lite

Getting Started with Transkribus Lite

Basic overview
Transkribus Lite
Last update 2 years ago
About Transkribus

Transkribus is a comprehensive solution for the digitisation, AI-powered text recognition, transcription and searching of historical documents. Find out more about Transkribus here

Transkribus is a comprehensive solution for the digitisation, AI-powered text recognition, transcription and searching of historical documents. Find out more about Transkribus here

Table of Contents

Table of Contents

This How-To-Guide gives an overview of the most important features of Transkribus Lite. This guide is supposed to offer you a basic understanding of the browser version of Transkribus and get your workflow up and running. 

Introduction 

Transkribus can be used for several purposes. The most important are:

  • Transcribe documents for a scholarly edition
  • Create training data to feed the Handwritten Text Recognition (HTR+ and PyLaia) system so it can learn to decipher your historical documents.
  • Run HTR on your documents and receive automatically generated transcripts.
  • Search for distinct words in your document collections with Keyword Spotting which is much more powerful than standard full-text search.
  • The platform lives from the community. The more data uploaded to Transkribus, the more efficient the program and especially the Handwritten Text Recognition will get

Transkribus lite is the web version of Transkribus with enhanced usability. Many of the beloved features from the Transkribus Expert Client can be used also in Transkribus lite. 

For more advanced features we also offer an Expert Client with the full functionality of the Transkribus platform. For more information about the Expert Client please visit this How-To-Guide.

Registration and pricing

To use Transkribus and Transkribus Lite you just need to register a free account on our website. Having an account is completely free. 

And also most of the features in Transkribus are free: Training, layout analysis, using Transkribus for manually transcribing, etc. Everything but – the very resource intensive – text recognition proper is free of charge. Yet, you can also try out the Transkribus text recognition feature for free, as you get 500 free credits (i.e. about 500 pages of HTR) upon signup.

Your purchases will enable us to keep Transkribus alive into the future and keep this wonderful community going. Everything you need to know about buying credits you find here.

Managing Collections

After you have logged into your account in Transkribus Lite, you will be redirected to the collection overview page. Here you can see and manage all the collections that you have access to. Read more about the management of collections here

Managing Documents

When you click on a collection in the collection overview page, the collection will open and you will see all the documents that are in the respective collection in the table tab. Additionally there is a search tab where you can search all your documents, a upload tab where you can upload new documents and a recognition tab where you can start Handwritten Text Recognition jobs.

Uploading files

In the “Upload” tab of the single collection view images as well as PDF files can be uploaded. The default uploader is the image uploader. Images in jpeg or png format can be uploaded with this upload option into a new document. The PDF upload option creates one new document for every PDF that you upload. Read more about uploading here

Using Handwritten Text Recognition

You can access the power of the automatic Handwritten Text Recognition with the Recognition tab. Here you can select the document you want to apply the HTR on, the HTR-model you want to use (one that you have trained or public ones) and the pages you want to recognize. Read more about HTR in Transkribus Lite here

Viewing pages

By clicking on “View” in the page overview you open the page viewer. By default you will see the uploaded document page on the left side. For a newly created document the right side will be empty. This is where the transcribed text will be displayed as soon as the document has been transcribed manually or with Handwritten Text Recognition

On the top right corner you will see the following action buttons:

Tags: Here you can hide, show and configure your tags

Show image and text

Show text only

Show image only

Edit page

[Version status | username] Version Selection

Transcribing and editing

By clicking on the Pen Icon in the single page view you can open the document editor. With the document editor you can edit your transcriptions. You can also use the editor for manual transcription. For this, you need to run a recognition job first in order to have the layout of the document analysed and the text regions and lines recognized. Don’t forget to save your work before you leave the document editor.

Searching documents 

On the search page you can search all of your collection. The Keyword(s) that you enter will be searched in all documents that are in collections that you have access to. For more detailed and enhanced search you can also use different search filters to narrow down the search results. The search results will display a preview of the text as well as a snippet of the image where the searched text was found.   

Viewing Jobs

The jobs page gives you an overview of all jobs that you have started. Jobs are tasks that Transkribus lite has executed on your behalf such as uploading documents or running Handwritten Text Recognition jobs. In the jobs list you will see the respective document the job was executed for as well as the type of job. You can also see the status and the duration of the job. By clicking on a job you will be redirected to the respective document.

Credits

The credit manager in Transkribus lite gives you full control over your credits. You can check the credits packages that you own, assign shareable credit packages to other collections and split credit packages into smaller packages (for instance, if you want to use one package in several collections). On the top of the credits manager you can see your overall credits balance i.e. the total of all credits that you have available in your packages. For more info check out our Credits How-To Guide