On April 27, 2023, Hassan Munshi, advised by Mark Liberman, successfully defended his PhD proposal, entitled "Arabic textsetting as sequence matching: A computational approach” (abstract below). His proposal committee included Jianjing Kuang and Gene Buckley and was chaired by Don Ringe.
Abstract: The influence of prosodic properties on textsetting varies across languages. While one language may predominantly rely on counting syllables, another may tend to strictly align stressed syllables with strong beats. This dissertation undertakes an exploration of the prosodic properties that are most relevant to textsetting, drawing on corpus and case studies from English and Arabic. Through the application of sequence matching using dynamic programming, a new model is proposed that aims to account for textsetting disparities across non-tonal languages. According to this approach, aligning text and tune can be viewed as an optimization problem, where a well-formed textsetting is the most optimal solution, based on the constraints imposed by the prosodic system of the language. Preliminary results indicate that lexical stress has the greatest impact on Arabic textsetting. Pending further results, this dissertation offers a new, computationally-efficient solution to the long-standing problem of textsetting.