Praat scripts
The excellent, multi-platform, free and open sound analysis / synthesis program Praat is written and maintained by Paul Boersma and David Weenink of the Institute of Phonetic Sciences at the University of Amsterdam, and can be downloaded from The scripts below should be called from within Praat and can be used and adapted freely under the terms of the GNU General Public License (GPL).

  • create harmonic signal [download] Creates a harmonic sound signal with a specified number of harmonics, and decay of harmonic amplitude.
  • create fm signal [download] Creates a harmonic sound signal with a specified time-frequency pattern.
  • print spectrograms multiple files [download] Prints (or creates eps files of) a list of spectrograms of sounds in different files. If the sound is too long it will be wrapped. Also useful for browsing the spectrograms of multiple sound files on screen.
  • read multiple files [download] Reads all files in a given directory with a specific extension. Handy if you want to load many sound files at once. This script works for the GNU/Linux version of Praat, but can easily be adapted to work on other platforms.
  • rms equalize [download] Equalizes the root-mean-square (rms) amplitude of a set of selected sound objects. The rms amplitude can be set to a value of choice, or (if left zero) will be chosen by the script so the maximum peak value in the set of sounds becomes +/- 1 (which is the maximum value defined in a wav file).
  • wiener entropy [download] Measures the Wiener entropy of a sound, which is a measure of its power spectral flatness. It is defined as the ratio of a power spectrum's geometric mean to its arithmetic mean, and expressed on a log scale. A completely flat power spectrum has a Wiener entropy of 0, and an infinitely narrow power spectrum has a Wiener entropy of minus infinity. This script chops up a sound in frames, and gives the mean entropy of all frames. Its execution is not efficient yet, because it creates and removes a lot of Objects. This will be improved when I have the time.