Centre for Language and Speech Technology


“An advanced rule-based unicode-aware tokenizer„

Ucto tokenizes text files: it separates words from punctuation, and splits sentences. It offers several other basic preprocessing steps such as changing case that you can all use to make your text suited for further processing such as indexing, part-of-speech tagging, or machine translation.

Ucto comes with tokenisation rules for several languages and can be easily extended to suit other languages. It has been incorporated for tokenizing Dutch text in Frog, our Dutch morpho-syntactic processor.


Ucto was written by Maarten van Gompel and Ko van der Sloot. Work on Ucto was funded by NWO, the Netherlands Organisation for Scientific Research, under the Implicit Linguistics project and the CLARIN-NL program.

The development and improvement of Ucto also relies on your bug reports, suggestions, and comments. Use the github issue tracker or mail lamasoftware (at) science.ru.nl.

Download & Installation

Ucto is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation.

To download and install Frog:

  1. First check if there are up-to-date packages included in your distribution's package manager. There are packages for Debian,Ubuntu and Arch Linux.
  2. If not, we strongly recommend you use our LaMachine software distribution, which includes Frog and all necessary dependencies, and runs on Linux, BSD and Mac OS X. It can also run as a virtual machine under any host OS.
  3. Alternatively, you can always download, compile and install Frog manually, as shown next.

Manual installation

To compile these manually consult the included INSTALL documents, you will need current versions of the following dependencies of our software:

As well as the following 3rd party dependencies:


The ucto documentation can be found here.

Python binding

Ucto can be used from Python through the python-ucto binding, which has to be obtained separately unless you are using LaMachine.