I am performing some large-corpus analyses in Praat and have been extracting formant (F1, F2, F3) values at every 10% interval for my textgrids, but have decided that a much more even-handed measure would be to extract these parameters once every half-second or so in each of my intervals, but I can't seem to figure out how to automate this. I am trying to pull relatively randomized formant samples from my intervals as the corpus is far too large to reasonably annotate each vowel.
Does anyone know how to script this in Praat (specifically just the time-series)? Thanks.
For reference this is how I've been sampling:
.time [1] = start + 0.1*(end-start)
.time [2] = start + 0.2*(end-start)
.time [3] = start + 0.3*(end-start)
.time [4] = start + 0.4*(end-start)
.time [5] = midpoint
....