Abstract: The dataset contains: Praat scripts for extracting and annotating relevant utterances from larger sound files, and extracting data (F0) from shorter sound files for further analysis. Sound files (.wav) containing single utterances Praat Pitch files with F0 contours of pitch accent tones Praat TextGrid Files R script for smoothing F0 contours using functional data analysis (fda), and making plots from and calculating correlation coefficients on the contours. All material from a corpus of 7 children engaging in free peer interaction and self reco...
(read more)
Topics: 
Linguistics
Speech recognition
Natural language processing