Abstract
The article presents the Mazovian Dialect Recordings Database, a digital resource documenting the dialectal variation of Polish. It outlines the methodological foundations of the database, rooted in fieldwork and in approaches to language documentation aimed at creating structured collections of spoken-language data. The analysis focuses on the structure of the database, which includes both archival and contemporary recordings, their chronological, geographical, and social variation, as well as the methods used for data processing and metadata development. Particular attention is paid to the integration of materials that differ with respect to their documentary status. The article also identifies possible applications of the database in linguistic research, teaching, and public outreach activities. The resource is presented both as an example of transforming archival material into an analytical tool and as part of the broader endeavour of providing digital documentation of spoken language.