This thesis provides tools for continuous and automatic
development of corpora from European parliamentary protocols.
The toolchain covers 22 EU member states and their 30
parliamentary chambers. It is the first set of tools of this
scope to this day.

While many corpora have already been built from parliamentary
data, this set of tools addresses the issue of being able to
collect and build such corpora with minimum human supervision
continuously, together with unified metadata.

These tools use official parliamentary websites of EU member
states as a source of reliable protocol data. After several
months of development, testing and usage, 30 corpora have been
built comprising altogether over 1 billion words and 60 million
sentences and the whole system continues crawling, processing
and building the respective parliamentary corpora on a daily
basis.