Braporus, spoken corpus of heritage russian in Brazil

Abstract

Heritage speakers represent a special category of bilinguals who are exposed to their first language at home in the childhood, but later acquire the main language of their society that becomes dominant. Brazil has numerous communities of heritage speakers of many languages such as Japanese, German, Italian, Polish, and Ukrainian; however, only few speech corpora are being collected. In the current work, we describe the protocol of the data collection and discuss some points about data management for the BraPoRus (Brazilian Portuguese-Russian) corpus, a spoken corpus of heritage Russian in Brazil. The participants are 26 elderly speakers who were born in Brazil or came to Brazil as children in the 1950s. The protocol of the data collection includes: 1) a brief sociodemographic questionnaire; 2) a working memory test in Russian and Brazilian Portuguese using the Month-Ordering task; 3) a semi-spontaneous narrative about the history of the participants’ family and their immigration to Brazil; 4) the Bilingual Language Profile; 5) a sociolinguistic interview with 139 questions; 6) unscripted dialogues between participants in Russian; 7) intonation task; and 8) reading task. The BraPoRus corpus contains more than 160 hours of speech recordings and represents a unique collection of heritage Russian in Brazil. We expect that the protocol described in this work will be useful both for Brazilian linguists who study other heritage languages, and for research on heritage Russian in other countries.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.