Search for a command to run...
In this paper, we introduce a new software program, Phon, which aims at providing unprecedented functionality for the elaboration, analysis and sharing of phonological data. Phon is part of a larger initiative, PhonBank, the brainchild of the first two authors of this paper, which aims at extending the software and database components of the CHILDES project into the realm of phonology. Phon can be downloaded from the CHILDES web site at: http://childes.psy.cmu.edu/Phon/. Phon is a new software program designed to support every step involved in the management, analysis and sharing of large corpora of phonological data. The foundation of this software was elaborated in the context of two Honors B.Sc. theses (Hedlund and O'Brien, 2004; Maddocks, 2005). Hedlund and O'Brien (2004) provided outlines and specifications of the main components of Phon. Maddocks (2005) contributed algorithms for segmental and syllabic alignment. This algorithm is an important component of the software, because it enables systematic comparisons between target and actual speech productions from both segmental and prosodic perspectives (see further below). Since May 2004, Gregory Hedlund and Philip O'Brien have continued working on the design and implementation of additional components of Phon with ongoing supervision from Byrne, Wareham and Rose, in close collaboration with MacWhinney's CHILDES-TalkBank team of programmers at Carnegie Mellon University. Despite the definite relationship between Phon and PhonBank, this paper will focus specifically on describing the current implementation of Phon, in section 2. We will also mention, in section 3, some future directions in the development of this project. We briefly conclude in section 4 with some of the potential impact that this software may have for the research community.