This repository contains the resources for the talk "Using machine-learning to analyse English vowels produced by Chinese learners of English" by Martin Schweinberger on April 11, 2024, at the University of Hildesheim.

The repository contains the following file types:

• Rmd: these files contain the R notebooks detailing the analytical steps of the study. These files can be downloaded and thus allow the local reproduction of the study.
• html: these files represent rendered R notebooks that can be viewed in any browser to see exactly what was done and why.
• rda: these files are located in the data folder and contain the sharable data sets used in this study.
• Rproj: the Rproj file is a so-called R project file which is a file manager neccessary when reproducing the analysis locally (on your own PC).

There are 3 Rmd and 3 html files which contain different steps of the analysis:

• Part 1 files contain the first part of the analysis where the individual vowel tokens were extracted.
• Part 2 files contain the data processing where the data sets were cleand and evaluated.
• Part 3 files contain the statistical analysis (only the Part 3 file is made available as an interactive Jupyter notebook which allows anyone to reproduce the statistical analysis via Binder.com).

Below is the link to interactive notebook that enable anyone to reproduce the analysis step-by-step or code-chunk by code-chunk via virtual machines launched at Binder.org

Interactive Statistical Analysis

This interactive notebook is located in a separate GitHub repository which is specifically set up to launch the interactive Jupyter notebook in a virtual machine via Binder.org.