A Corpus for Large-Scale Phonetic Typology

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4526–4546 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 4526 A Corpus for Large-Scale Phonetic Typology Elizabeth SaleskyC Eleanor Chodroff7 Tiago PimentelZ Matthew WiesnerC Ryan CotterellZ,D Alan W Blackì Jason EisnerC CJohns Hopkins University 7University of York ZUniversity of Cambridge DETH Z¨urich ìCarnegie Mellon University esalesky@jhu.edu eleanor.chodroff@york.ac.uk Abstract A major hurdle in data-driven research on ty- pology is having sufficient data in many lan- guages to draw meaningful conclusions. We present VoxClamantis V1.0, the first large- scale corpus for phonetic typology, with aligned segments and estimated phoneme- level labels in 690 readings spanning 635 languages, along with acoustic-phonetic mea- sures of vowels and sibilants. Access to such data can greatly facilitate investigation of phonetic typology at a large scale and across many languages. However, it is non- trivial and computationally intensive to ob- tain such alignments for hundreds of lan- guages, many of which have few to no re- sources presently available. We describe the methodology to create our corpus, discuss caveats with current methods and their impact on the utility of this data, and illustrate pos- sible research directions through a series of case studies on the 48 highest-quality read- ings. Our corpus and scripts are publicly available for non-commercial use at https:// voxclamantisproject.github.io. 1 Introduction Understanding the range and limits of cross- linguistic variation is fundamental to the scientific study of language. In speech and particularly phonetic typology, this involves exploring po- tentially universal tendencies that shape sound systems and govern phonetic structure. Such investigation requires access to large amounts of cross-linguistic data. Previous cross-linguistic phonetic studies have been limited to a small number of languages with available data (Disner, 1983; Cho and Ladefoged, 1999), or have relied on previously reported measures from many studies (Whalen and Levitt, 1995; Becker-Kristal, 2010; Gordon and Roettger, 2017; Chodroff et al., 2019). 8765 9 MCD Figure 1: The 635 languages of our corpus geo-located with mean Mel Cepstral Distortion (MCD) scores. Existing multilingual speech corpora have similar restrictions, with data too limited for many tasks (Engstrand and Cunningham-Andersson, 1988; Ladefoged and Maddieson, 2007) or approximately 20 to 30 recorded languages (Ardila et al., 2020; Harper, 2011; Schultz, 2002). The recently developed CMU Wilderness corpus (Black, 2019) constitutes an exception to this rule with over 600 languages. This makes it the largest and most typologically diverse speech corpus to date. In addition to its coverage, the CMU Wilderness corpus is unique in two additional aspects: cleanly recorded, read speech exists for all languages in the corpus, and the same content (modulo translation) exists across all languages. However, this massively multilingual speech corpus is challenging to work with directly. Copy- right, computational restrictions, and sheer size limit its accessibility. Due to copyright restrictions, the audio cannot be directly downloaded with the sentence and phoneme alignments. A researcher would need to download original audio MP3 and text through links to bible.is, then segment these with speech-to-text sentence alignments distributed …

Content truncated to 500 words