In this paper we discuss some theoretical and practical problems of the collection and presentation of spoken data for dialectological purposes. We will pose questions connected with audio data recording, digital audio files storage, data transcription, transcription quality checking. Furthermore, we discuss pros and cons of the application of morphosyntactic tagging models to our data. More specifically, the paper describes the main activities and projects of the Centre for the documentation of dialects of Southern Lazio, set up by the authors at the University of Cassino.