Attached is a revision to Nick’s script for gathering data. Thanks to Nick for laying the foundation and writing the code on the formulas.
This script allows you to search for a particular vowel (e.g., AE1, UH1, etc.) or a particular word (e.g., foot, rat, etc.). So you do need to know the vowel you’re looking for either way and whether or not that vowel is in your target word. Eric is working a table of Arthur which we will be able to use as a cheat sheet. The other changes I made was to have the bark difference values recorded in the spreadsheet, whether the trajectory was rising, falling, fronting, backing and the context (pre/post sounds) and the word in which the sound occurs, as shown below.
Here are the instructions you’ll need to know to run the script successfully.
1. open the corresponding .wav and .TextGrid files into the Praat object window (Open > Read from file…)
2. open the script (Praat > Open Praat script…)
3. select the .TextGrid file in the object window
4. go over to the script window, and find line 48 (Search > Go to line…). If you are going to collect a bunch of subjects and want to save yourself some time, I’d suggest changing the line,
fappendinfo ‘objName$’_data.txt
to something that will be less dynamic (as it is now because it passes the name of the sound file into the data file name) consistent for all files, such as
fappendinfo my_data.txt
5. once you’ve saved the changes to the script (File > Save), run the script (Run > Run). You should see the following window:
6. change the nulls to values:
a) if you want to run the script for all of one vowel (e.g., UH1), then put this in and press OK.
b) if you want to run the script for a specific word, then put this word in the null place IN ADDITION TO the vowel, as shown below.
This will be saved to your file which can be opened up in Excel.
Please let me know if you encounter an error and it’ll get fixed.