This article consists of two parts: an introduction to the principles and the current state of the corpus of Estonian dialects, and a presentation of the main characteristics of the vowel systems of Estonian dialects based on statistical analysis of the data in the dialect corpus. First, the starting points and problems that had to be taken into consideration when compiling the corpus are introduced, and the development of the project up to now reviewed. Thereafter, the state of the dialect corpus as it stands in October 2003 will be described together with the principles of tagging and a frequency study of the dialect vocabulary carried out on the basis of the corpus. The characterization of the vowel systems of Estonian dialects will be presented according to the general distribution of the distinctive features.