VowSpace is an open-source desktop application designed with the aim of acquiring, visualizing, normalizing, and linguistically analyzing vowel sounds from audio files and datasets.
Using VowSpace, you can visualise a dataset or compare datasets from different sources—whether across languages or individuals (e.g., two different languages, three idiolects).
I developed this program as a creative diversion while getting one of my phonetics homework done. The task involved manually inputting data into Excel and adhering to specific guidelines to construct a scatterplot resembling a vowel space.
I've provided two different codes. The first script is a standalone Python application enabling you to generate a scatterplot or compare different vowel spaces by inputting or importing data. The second programme (in the Vsv_Python_R folder), consists of a Python interface that allows you to enter data and save the entered data as a .csv document. Subsequently, an R script reads this dataset, generating a scatterplot based on the provided information according to some specifications.
-
Reading Data from External Sources: Read data from any .xlsx file.
-
Data Input: Easily input new data by pressing the "Add Data" button or using the "Enter" key.
-
Save Scatterplots: Save the visualised data as a scatterplot as JPEG or PNG using the "Save Scatterplot" function or the "Ctrl (Command) + S" shortcut.
-
Saving and Manipulating Data: Save a dataset that you worked on as an .xlsx file, also edit any dataframe using the DfEditor.
-
Data Normalisation: You can use the "Data Settings" to normalise the data using various normalisation methods.
-
Scatterplot Settings: You can edit the visuals on the scatterplot using the "Visualization Settings" menu.
-
Audio Analysis Tools: Show pitch, intensity and/or formants of an audio file and add them to the main window.
Under the "Release" section, you can download the application in .app format for Microsoft Windows and Apple Macintosh devices.
Or
-
Clone the repository to your local machine.
-
If you are importing data from a document, it needs to have columns labelled "F1", "F2" and "speaker".
git clone https://github.com/alicagank/vowspace.gitGNU General Public License v3.0
This application was developed by Ali Çağan Kaya as part of a linguistic project at Hacettepe University, Turkey. If you find this tool useful and decide to modify or publish it, kindly give credit to the original author.