Rhythm Analysis

A Shiny Interface for computation and visualization of rhythm metrics

This app provides an interface for analyzing rhythmic metrics (such as %V, varcos, PVI) of speech based on annotated TextGrid files. It allows you to compute a variety of standard rhythm metrics and visualize them directly within the app.

👉 Access the app via the UB Phonetics Lab Note: This app is hosted on a UB server. We do not save the data but if your files contain personal, private, or sensitive data, you may prefer downloading the code and run it locally using RStudio with the shiny package, or deploy it to a secure server managed by your institution.

Input Requirements

• At least two Praat TextGrid files (or more), containing labeled intervals for consonants and vowels.
• Encoding: UTF-16 preferred.
• The transcription can be located in any tier — just specify which tier to use.
• Compatible transcription formats include:
  • Orthographic
  • Phonetic
  • PTK / a
  • C / V

Output

• A set of numeric rhythm metrics, including:
  • Speech rate (sounds/second)
  • %V (percentage of vocalic intervals)
  • %C (percentage of consonantal intervals)
  • VarcoV / VarcoC
  • DeltaV / DeltaC
• Visualizations:
  • Interactive plots of duration data and rhythm metrics, PCA, MDS and dendrograms.
• Downloadable data:
  • Duration database
  • Computed rhythm metrics as a .csv file

📌 Citation: If you use this app in your research, please cite:
Elvira-García, Wendy. (2025). Rhythm Analysis: A Shiny Interface for computation and visualization of rhythm metrics. [DOI or URL.] (https://doi.org/10.5281/zenodo.15401985)