The VinKo Corpus. Oral data from Romance and Germanic local varieties of Northern Italy

The VinKo corpus is a parallel corpus with audio recordings from German and Italian dialects and minority languages spoken in the Italian regions Trentino-South Tyrol and Veneto. The data has been crowdsourced via the online platform of the VinKo project and was produced in response to a pronunciation and translation task targeted at eliciting phonological and morpho-syntactic phenomena for language contact studies. The VinKo corpus V1.1 contains over 125.000 audio files from 11 language varieties. The project strives towards a ‘open science’ approach with an integral ‘citizen science’ component by active collaboration with local institutions and freely sharing the data with different stakeholders, e.g. speech communities, scientific community. All collected data can be accessed via the admin interface of the VinKo website or downloaded from the online repository, and a selection of the data is represented via an online map targeted at a non-specialist audience.