Abstract
The article provides a view of special corpora being created within the Czech Language Acquisition Corpora project ROMi (Czech language of Romany speakers) representing spoken and written language of Romeny learnes aged 6-28 years, supplemented with information about learners and texts. The article also describes methods of collecting and processing the language data as well as brief list of main errors made by Romany users of Czech language.
There is also demonstration of texts written by Romany users of Czech.