Current questions about language, its use, acquisition, and related disorders can nowadays be tested on empirical grounds that are firmer than ever. Among other developments, the democratization of computer-assisted approaches to linguistic analysis and the fostering of more forward-thinking attitudes toward data sharing have led to new and exciting research opportunities. These advances, and the outcomes they yield, can in turn be used to feed more traditional but still fundamental debates in theoretical linguistics. Modern approaches to corpus-based linguistic studies call for robust systems for databasing and analyzing broad corpora of linguistic data.
The CHILDES project (Child Language Data Exchange System; [url]http://childes.psy.cmu.edu/[/url]) has, over the last 25 years, provided the grounds for thousands of publications in areas ranging from language acquisition to computational linguistics, and provides a model example of this. The PhonBank project broadens the scope of the CHILDES system to include the analysis of phonological development in first and subsequent languages for learners with and without language disorders.
Since its inception in 2006, PhonBank has contributed computational and databasing facilities for corpus-based studies of phonological development, many of which can also be used in other areas of corpus phonology, including fieldwork and multilingual studies. Using these tools, researchers can perform systematic investigations based on large, fully transcribed corpora of speech data, a number of which also incorporate audio or video recordings. At the centre of these facilities is Phon, a software program that greatly facilitates a number of tasks required for the analysis of phonological development. Phon, whose development takes place at Memorial University of Newfoundland supports multimedia data linkage, unit segmentation, multiple-blind transcription, automatic labeling of data, and systematic comparisons between target (model) and actual (produced) phonological forms.
