This is the source for my dissertation at the University of Rochester ("Perception in a Variable but Structured World: The Case of Speech Perception"). It's a bit of a Frankenstein's monster, cobbled together from Latex files from previously published papers (Chapters 2 and 3) and Rmd files for some new, as-of-yet unpublished work (Chapters 4 and 5). The whole thing is held together with pandoc and makefiles.
To download a PDF, see the releases.
I hope there might be some use in seeing how I managed to cobble together a single, automatically built document from heterogeneously formatted source files. The commit history may be informative so you can see the process of converting a stand-alone Latex file to one that plays nicely with pandoc.
Also, in the spirit of openness, you might like to see how the figures and results from the later chapters were generated.
Unfortunately, you'll likely not be able to build it yourself, since the Rmd files require some data that was shared privately by others and is not publicly available. Contact me if you want access to something and we'll see what we can do.
A (possibly non-exhaustive) list of dependencies is:
pandoc,pdflatex, andmakefor converting things to latex, combining them, and rendering the final PDF.nspvowels: for Chapter 5, an R package with a small collection of vowel F1/F2 values from the Nationwide Speech Project, and code to analyze it (I'm planning to split the code and data out separately).votcorpora: an R package with aggregated collection of VOT values for voiced and voiceless word-intial stops (donated by a number of other researchers).beliefupdatr: an R package with helpers for coding phonetic adaptation.daver: an R package with misc. functions I've found useful.