Praat Code

Here are a few useful Praat scripts (and packages) I’ve written / adapted from others. For a full list, see my Github: https://github.com/michellecohn/praat-scripts/

Loading files

Load in all .wav files

Load in all .wav files / associated TextGrids

Save all .wav and/or .TextGrid files to a directory

Converting

Convert all files to mono

Convert to stereo

Convert from .mp3 to .wav

Override sampling frequency

Convert to short Text Grid

Resample all files & combine with other sound

Remove additional tiers (prepare for MFA)

Adjusting acoustic properties

Change mean intensity of all files

Match intensity contour of another sound

Match vowel duration of another sound

Match mean dB of another vowel

Match f0, intensity, & duration of another sound (combined script)

Create duration continuum

New monotone f0 for the whole sound

New monotone f0 for selected vowels

New mean f0 for sounds from a list (.txt file)

Pull in all .mp3 files (options to amplitude normalize, resample) and save as .wav files

Adding silence

Pad beginning of the sound with silence

Pad end of soundfile so that total duration is the same value (e.g., 900 ms)

Concatenate sounds with a pause

Add buffer to soundfile AND shift textgrid accordingly (currently adds 1/2 total duration to beginning)

Adding noise

Pad end of sound with 1 second of white noise (can adjust duration and intensity of noise)

Gating Package (MC 2/2024)

This is a Praat package for taking words from a TextGrid, removing the coda, and gating into white noise. You will need to specify an input directory (.wav + .TextGrids) as well as an output directory. The scripts will create intermediate folders in that output directory.

• Can ramp into noise based on a percentage of the overall vowel duration or a set durationLoading / saving into Praat

Speed up or slow down a soundfile + textgrid

Multiply soundfile and textgrid by a duration factor

Speed up / slow down textgrid (e.g. from MFA output) to match original soundfile

Annotations

At each “sp”, label separate sentence interval (e.g., from FAVE align output)

Mark silences (no context; i.e., no FAVE or MFA)

Get sentence from labeled word intervals (assumes you have (unlabeled) sentences in separate intervals from Above step)

Measurements

Get intensity, duration, & mean f0 over 12 points of the vowel (Adapted from Christian DiCanio)

Get duration measurements for each annotated word

Get duration of vowel and coda for each word

Combining sounds

Concatenate 2 sounds (1 sound == frame)

Concatenate all sounds in a directory

Resample all files & combine with other sound

Extract & save all sounds

… from a force aligned file (finding “sp”s) and numbering sentences

Split individual sounds from a word and save to directory

Writing to .txt file

Extract info before underscore (to command window)

Concatenating all files in a particular condition

Open all files in a directory & concatenate them, saving a .txt file transcript

Pure tone generation

Create 3 tone headphone screen (Woods et al., 2017)

Combining multiple textgrids

Combine MFA + other textgrid into one massive textgrid