Articulatory Data Anaysis

ADA is a free and open-source tool for articulatory data analysis. It a software for the visualization, annotation, and post-processing of data collected via electromagnetic articulography (EMA) as well as capabilities for post-processing and conducting measurements. ADA utilizes a graphical user interface and is especially designed to make the work with this kind of data easily without requiring programming skills from the users.

ADA is currently work-in-progress and has multiple features already implemented (see below). Fell free to try the software!

Features

In its current state, ADA allows the import and export of and into various formats for EMA data, audio and annotation files. EMA data can be visualized in various ways, and articulatory landmarks can be set either manually or automatically. Furthermore, ADA allows the extraction of parametric and dynamic measurements.

See below a list of features that are currently implemented.

  1. Import formats
  • EMA: pos (AG500/AG501), csv, netcdf
  • Audio: wav, ogg, mp3
  • Annotation: Praat TextGrids, lab, json
  1. Visualization
  • Visualization of sensor positions over time, as well as their velocity, tangential velocity, and Euclidean Distances
  • 2D visualation of sensor positions
  1. Annotation
  • Manual annotation
  • Automatic annotation of landmarks using various methods (based on velocity or tangential velocity) and thresholds (15%, 20%)
  • Automatic annotation across files
  1. Measurements
  • Parametric measurements: sensor positions, velocity and time at specific landmarks
  • Dynamic measurements: extraction of sensor positions in specific ranges (based on landmarks or acosutic segment boundaries)
  1. Export formats
  • EMA: csv, netcdf
  • Annotation: Praat TextGrids, csv, json

Dependencies

Future work

  • Writing up the documentation
  • Improving perfomance
  • Support for data from AG100/AG200 and NDI WAVE
  • Implementation of a head-correction

Contact

If you have any suggestions or if you want to report bugs, you can either open an issue with the appropriate label in the GitHub repository here or contact me via

name (point) surname (at) sorbonne-nouvelle.fr

Citation

If you want to cite ADA in your publications, please cite the GitHub repository:

Buech, P. (2024). ADA - a tool for articulatory data analysis. GitHub. Retrieved, 04 May 2024 from https://github.com/phbuech/adatool

Acknowledgements

This work benefited partially from a government grant managed by the ANR under the ”Investissements d’Avenir” programme with the reference ANR-10-LABX-0083. It contributes to the IdEx University of Paris - ANR-18-IDEX-0001 (LABEX-EFL). It was further partially funded by XYZ.