Preface

This document was created as accompanying text for a workshop called “Analysis of fundamental frequency trajectories using generalized additive models” presented on the “6th International Symposium on Applied Phonetics”—ISAPh2026 held in Hiroshima, Japan, on September 14-16, 2026.

In this workshop, we introduce GAMs for the analysis of fundamental frequency trajectories. Because of time limitations, this introduction should serve as a starting point to inspire further exploration by interested researchers. We cover data preparation, common pitfalls, simple modelling, diagnostics, prediction, and reporting. Although the workshop uses fundamental frequency trajectories, the same techniques could be used for the analysis of other time-related phonetic phenomena that unfold over time, frequency or space, such as trajectories of formants, open/close quotients (from electroglottographic recordings), long-term average spectra, or tongue contours (from ultrasound recordings).

The dataset for the practical example is freely available along with this tutorial from https://julovi.codeberg.page/gam_tutorial. Prepared with bookdown, the material can be browsed as a webpage, e-book, or PDF.

The workshop assumes some knowledge of R (R Core Team 2018) and basic statistics. If this is not the case, the reader is welcomed to visit (Villegas 2018)

Creative Commons License
This material is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. It was created based on A Minimal Book Example by Yihui Xie.

References

R Core Team. 2018. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. https://www.R-project.org/.
Villegas, Julián. 2018. “Introduction to statistical analysis of phonetic data in R.” 2nd Int. Symp. On Applied Phonetics (ISAPh2018) (University of Aizu), September.