Skip to content

wantingyeh/Praat

Repository files navigation

Praat

create-textgrid-forall-DIME

  1. Create textgrids for all wav files in the same directory.
  2. Automatically detect the space and the sounding part
  3. Generate 3 tiers: (1) parent's ID, (2)Infant's ID, and (3) silence and sounding parts

Noise deduction

  1. Use spectral substraction method to denoise the ambient noise
  2. Parameters: window length 0.025 s, filter frequency range: 50 - 120000
  3. Smoothing: 40 Hz
  4. Reference for the used parameters: S.F. Boll (1979): "Suppression of acoustic noise in speech using spectral subtraction."IEEE Transactions on ASSP 27: 113–120.

Intensity

  1. Scales the intensity of all the files in the same directory

save_intervals_to_wav_sound_files_text

  1. This script saves each interval in the selected IntervalTier of a TextGrid to a separate WAV sound file.
  2. The source sound must be a LongSound object, and both the TextGrid and the LongSound must have identical names and they have to be selected before running the script.
  3. Files are named with running index numbers.
  4. A text file is created from the interval labels. Each line in the text file refers to a sound file
  5. This script is distributed under the GNU General Public License. (by 8.3.2002 Mietta Lennes)

save-to-PitchFile-PitchInfo

  1. automatically save .Pitch and .PitchTier from a folder (only search for wav files)
  2. extract frequency info: F0min, F0max, F0mean, F0std, slope, in Hertz and semitones

BatchConvertTxt.xlsm

  1. A vba script from excel
  2. Convert all the textgrids to txt files in one directory, so that we can use the txt files to put in excel file for further coding

About

Praat scripts for speech analysis

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published