COALA voice data and transcripts Italian

This dataset contains audio files and transcripts in Italian and related to manufacturing. We collected the scripts during the Horizon Europe RIA COALA (GA 957296, project reference website) from industrial use cases and hired a service provider to generate the related audio files (BIBA - Bremer Institut für Produktion und Logistik GmbH ordered the service). The service provider checked the audio files for quality.

The service provider recruited crowd workers, and gathered their audio records, informed consent (privacy) and agreement that their records become public domain (Creative Commons 0; https://creativecommons.org/share-your-work/public-domain/cc0/). The service provider declared to follow a Crowd Code of Ethics and a Fair Pay policy.

The metadata file contains the following information:

file_name: name of the audio file

script: script the speaker had to speak

scriptId: the numeric identifier of the script

participantId: the numeric identifier of the participant (speaker)

gender: the gender as indicated by the participant (MALE or FEMALE)

age: the age in years as indicated by the participant

age_range: the age range in years  (18-30, 31-45, 46+)

country: the birth country indicated by the participant

current_country: the country of residence indicated by the participant

primary_language: the language indicated as primary by the participant

ever_worked_factory: answer to the question: "Have you ever worked in a factory, manufacturing setting?" (Yes/No)

years_worked_factory: answer to the question: "If yes, for how many years?" (1-10, 10+)

background_noise_type: background noise in the audio as indicated by the participant (mild, humming/technical, no noise)

gdpr_and_ipr_consent: answer to the privacy notice and the ipr transfer to CC-0 (Yes)

date_signed: date when the participant signed the consent form (US format, MM.DD.YYYY)