KIParla - KIPasti transcripts

The KIPasti corpus is part of the larger KIParla collection (www.kiparla.it), which can be freely queried through the NoSketch Engine interface. The ParlaBO corpus was compiled within the framework of “DiverSIta – Diversity in spoken Italian” project, funded by the Italian Ministry of University and Research (MUR) (PRIN 2022 PNRR Call). It consists of over 40 hours of spoken data collected in thirteen different Italian regions (Abruzzo, Basilicata, Calabria, Campania, Emilia-Romagna, Lazio, Lombardy, Marche, Apulia, Sardinia, Tuscany, Umbria, Veneto) during mealtime conversations, generally within family settings. The interactions, recorded between 2020 and 2024, involved 145 speakers with different origins, ages, education levels, and occupations. Italian is predominantly used in all interactions, but in most of them (78%), various passages in dialect are also present. The transcriptions have been anonymized. Overall, the module is made up of 63 conversations. This repository contains: - metadata for both speakers (occupation, gender, age, origin, L1, educational achievement) and conversations (collection point, year, languages used), in the metadata subfolder - descriptions of the set of transcription conventions used for this module - for each conversation you will find: .eaf file in eaf/ folder (time-aligned Jefferson-style transcriptions); .txt file in linear-jefferson/ folder (linearized Jefferson-style transcription); .txt file in linear-orthographic/ folder (linearized transcription retaining only orthographic words); .tsv file in tsv/ folder (tokenised version of the transcription). More information can be found in the README.md file. Due to GDPR restrictions, pseudo-anonymized audio files (MP3) are available under a restricted-access license. To request access, please contact the corpus coordinators through the KIParla website and follow the provided procedure. This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.