Sam Passmore, Lila San Roque, Kirsty Gillespie, Saurabh Nath, Kira Davey, Keira Mullan, Tim Cawley, Jennifer Biggs, Rosey Billington, Bethwyn Evans, Nick Thieberger, Danielle Barth. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

For most of the world’s languages, detailed phonetic analyses across different aspects of the sound system do not exist, due in part to limitations in available speech data and tools for efficiently processing such data for low-resource languages. Archival language documentation collections offer opportunities to extend the scope and scale of phonetic research on low-resource languages, and developments in methods for automatic recognition and alignment of speech facilitate the preparation of phonetic corpora based on these collections. We present a case study applying speech modelling and forced alignment methods to narrative data for Nafsan, an Oceanic language of central Vanuatu. We examine the accuracy of the forced-aligned phonetic labelling based on limited speech data used in the modelling process, and compare acoustic and durational measures of 17,851 vowel tokens for 11 speakers with previous experimental phonetic data for Nafsan. Results point to the suitability of archival data for large-scale studies of phonetic variation in low-resource languages, and also suggest that this approach can feasibly be used as a starting point in expanding to phonetic comparisons across closely-related Oceanic languages.

We present an exploratory analysis of several long-term distributional measures of f0 range in the speech of university-educated speakers of Indian English from four L1 backgrounds (Telugu, Tamil, Hindi and Bengali). The aim of this study is to investigate the degree of homogeneity in Indian English prosody and any similarities between the speakers' productions in English and their L1. Following recent studies, we examine three aspects of f0 range: pitch level (relative height of habitual f0), pitch span and pitch dynamism. Overall, across varieties, pitch level measures reveal individual speaker differences and only weak L1 effects on max f0 and median f0. Some speakers show higher f0 in their L1 productions compared to their English productions. More robust patterns were found for pitch span and dynamism: for all measures (maximum-minimum f0, pitch dynamism quotient and standard deviation), significant differences were found between L1 and English (p<0.001) for Bengali and Telugu L1 speakers. The relative weakness of L1 effects would suggest a degree of homogeneity in Indian English, at least for the prosodic parameters investigated. Evidence of a shift in pitch span when talking in English, regardless of L1, further suggests a convergent speech variety.