Building Infrastructure for Longitudinal Speech Data Collection

点击翻译按钮获取中文摘要

Recordings of natural speech are crucial for research on a wide-range of topics in the language sciences. Audio diaries can form an important source for data, especially for research at the interface of language and society. Audio recording allows diarists to express themselves in ordinary and casual ways, giving language researchers a clearer view of everyday language in the past and the present. More generally, diaries have long been important tools in human history for recording events and experiences, providing information that can contribute to the understanding of cultural heritage. This project builds on a pilot study that began in April 2020, capturing people's day-to-day experience of the COVID pandemic and beyond. Diarists use a custom mobile app to record their spoken diary entry. Mobile apps democratize data collection by making it easy for anyone to contribute as a citizen scientist. App code is made public, for other scientists to adapt for their own use. Anonymized audio and transcripts are made available to the public and other researchers in perpetuity via a public library repository. This award supports the improvement and expansion of the remote audio diaries infrastructure to optimize participant recruitment and retention through the development of improved app technology and long-term community partnerships with youth organizations. These self-recorded audio diaries provide several benefits for linguistic research. First, researchers can obtain data from participants across a wide geographic area, enabling a large-scale analysis of ongoing language changes. Second, repeated entries from the same speakers enable researchers to track individual participation in community-level sound change in real time. Third, leveraging the genre of diary entries results in important personal social insights from speakers, enriching the understanding of the relationship between linguistic and social processes. The data collected enable researchers to test existing hypotheses about preteen sociolinguistic calibration, by comparing variability of ongoing sound changes in child, preteen, and teenage speakers' language production. Existing hypotheses regarding post-adolescent participation in ongoing sound change can also be tested. The success of this infrastructure will facilitate continuous collection of data beyond the requested funding period.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.