This is a Praat plug-in which segments TextGrid annotations into segments, syllables or words. The resulting segmentation is stored as a new tier whithin the same TextGrid.
- Praat 6.0.40 or later (download here)
In the Object window, go to Praat > Tokenize > Settings..., a dialog box will appear.
Once you are done with the settings, you are ready to process your own TextGrids. Before starting, remember that you need to provide an interval tier as an input for this plug-in to work. Here is an example.
With this plug-in, you can segment TextGrids in the Object window and those stored in a folder.
First, select those TextGrids that you want to segment. Then, go to TextGrid: Add tokenized tier...
When you click on it, you will see a dialog box. In Input tier, write the name of the tier where your annotations are stored.
Then, check the segmentation levels to be be generated. Finally, press on Apply or Ok. The TexGrids are now segmented!
In the Praat menu, go to Praat > Tokenizer > Tokenize(do all)...
You will see a dialog box similar to one shown in the previous case.
In the Folder with annotation files put the directory where your TextGrids are located in your machine.
In Save results in, copy the path where the resulting files will be stored.
Then, complete the other fields as explained before and press on Apply or Ok. The resulting files should be in the destiny directory.
- Rolando Muñoz Aramburú
This project is licensed under the GNU GPL terms - see the LICENSE.md file for details.
Muñoz A., Rolando (2018). Tokenizer[Praat plug-in]. Version 1.2.0, retrived 20 Sep 2018 from https://gitlab.com/praat_plugins_rma/plugin_tokenizer