ML p(r)ior | Automated tone transcription

Automated tone transcription

9410022 | cmp-lg
In this paper I report on an investigation into the problem of assigning tones to pitch contours. The proposed model is intended to serve as a tool for phonologists working on instrumentally obtained pitch data from tone languages. Motivation and exemplification for the model is provided by data taken from my fieldwork on Bamileke Dschang (Cameroon). Following recent work by Liberman and others, I provide a parametrised F_0 prediction function P which generates F_0 values from a tone sequence, and I explore the asymptotic behaviour of downstep. Next, I observe that transcribing a sequence X of pitch (i.e. F_0) values amounts to finding a tone sequence T such that P(T) {}~= X. This is a combinatorial optimisation problem, for which two non-deterministic search techniques are provided: a genetic algorithm and a simulated annealing algorithm. Finally, two implementations---one for each technique---are described and then compared using both artificial and real data for sequences of up to 20 tones. These programs can be adapted to other tone languages by adjusting the F_0 prediction function.

Highlights - Most important sentences from the article

Login to like/save this paper, take notes and configure your recommendations