Digital processing of dialectal texts published in Chisinau in the years 1969–1987

Abstract

The dialectal texts published in Chisinau in the years ‘60–‘80 of the past century are of particular importance for research in different fields of science: linguistics (especially dialectology, the history of the Romanian language, stylistics), history, ethnography, folklore studies, sociology, psychology, ethnolinguistics, sociolinguistics, psycholinguistics. Revitalizing them through digital processing and transliteration from the Cyrillic script into the Latin script will allow their full exploration. Currently, from the six volumes of the collection Dialectal Texts: Supplement to the Moldavian Linguistic Atlas, the first volume has been scanned, recognized, and transliterated resulting in the Latin script text. A set of the accompanying items (recognition templates, the user alphabet, and the spelling dictionary) have been prepared to automate the process. Based on the established transliteration rules, a converter was developed to convert the recognized text of the Romanian local speech samples from the Cyrillic-based phonetic transcription into the Latin-based one.