The existing corpora of Contemporary Bulgarian Language are intended to analyze the morphological and syntactic features of the language’s oral form. The description of the phonetic level is the least studied. The created database of Bulgarian Speech with the program SARP represents an initiative to fill this lack.