This is a tool to help comparing subsets of speech WAVs. It is using opensmile for feature extraction and RStudio Shiny for presenting the statistics.

Installation

1. Install R
2. Change to the folder of this project
3. Start R and install renv and dependencies

   install.packages("renv")
   renv::restore()

Using

1. Prepare files:

   • Pack WAVs with ZIP

   • Create CSV file with factors.

     It should have a line for each WAV file and file name should be in the first column with name file. Ex:

     file,f1,f2
     003.wav,calm,male
     004.wav,energetic,female
     

2. Run the server:

   Rscript app.R

3. Use browser http://localhost:8008