Analyze_parameters

This script helps you to extract acoustic values of the speech file easily. You need to have sound files and their label files.  The format of the output file is CSV, which is comma-separated version, and the output file contains the name of the files, analyzed label, and acoustic values of fundamental frequency, first formant, second formant, third formant, and duration of target label.

Right click the link and save it.

Followings are instructions to used the script.

When you open and run the script with Praat, you will see the following screen.

You can set the parameters in this screen as you want.

  1. Directory of sound files: Type the location of the directory that contains the sound files. You can type absolute path (begins with “C:/” in Windows, “/User/” in Mac, and “/home/username/” in Linux) or relative path (begins with “./” or “../”).
  2. Sound file extension: Type the extension of the sound files. All sound files should have same extension.
  3. Directory of TextGrid files: Type the location of the directory that contains the TextGrid files. TextGrid files should have labeling information.
  4. TextGrid file extension: Type the extension of the TextGrid files.
  5. Full path of the result: type the path, name, and extension of the result file you want to create. Absolute and relative path both work.
  6. Which tier ~: Type the name of the tier you want to analyze. This script will extract acoustic parameters of each label in the designated tier.
  7. Formant analysis parameters: Parameters which will be used to extract the formant values.
  8. Pitch analysis parameters: Parameters which will be used to extract the fundamental frequency.

After typing all parameters, press Apply or OK button. If you press Apply, the buttons will become clickable when the output file is generated. If you press OK button, the screen with parameters will disappear.

The output file will look like as follow:

Since you can open CSV file with Excel, you can easily modify the output file. If you want to analyze the result statistically, you can read this file with R, and do statistical analysis.

 

Seongjin Park

Categories: LinguisticsPhonetics

Leave a Reply

Your email address will not be published. Required fields are marked *