Praat++ is a browser-based multimodal annotation platform built on Praat functionality.
It provides real-time waveform, spectrogram, pitch, and intensity visualization, supporting annotation of both acoustic and video data directly in the browser with features like region labeling, tier control, and auto-saving.
- Duplicate file detection: Uploading a file with the same name will trigger a warning.
- Multiple file upload: Select and upload multiple
.wav.mp3or.mp4files at once.
- Region create: Drag on the waveform to create a new annotation region.
- Region update: Resize or move regions freely.
- Playback: Click a region to play the corresponding audio.
- Timestamp preview: Hover over the waveform to view precise timing.
- Annotation: Right-click a region to add or delete labels.
- Select tier: Click a tier to activate it before annotating.
- Edit tier name: Click the tier name to edit it directly.
- Hide/delete: Right-click to hide or delete a tier.
- Scroll to zoom: Use the mouse wheel over the waveform to zoom in or out.
- Fully customizable: Define your own annotation labels and checkbox attributes.
- All changes are auto-saved: No need for a Save button.
- Export to TextGrid: Download annotations as a
.TextGridfile compatible with Praat.
To explore the full features of Praat++, please refer to the Praat++ Tutorial.
All annotation data and files generated during processing are stored per file per user under the following structure:
static/videos/pool/{user}/{file_name}
Each uploaded file is saved under the logged-in username, and annotations persist independently per user.
You can run Praat++ locally by following these steps:
git clone https://github.com/UTA-ACL2/PraatPlusPlus.git
cd PraatPlusPluspython -m venv venv
# On Windows:
venv\Scripts\activate
# On macOS/Linux:
source venv/bin/activatepip install -r requirements.txt- Install
ffmpegand ensure it is added to your systemPATH
Open login_routes.py and define allowed usernames in VALID_ACCOUNTS.
python run.pyVisit http://127.0.0.1:5000 to use it.
Peter (Weiran Zhang)
Email: [email protected]
This tool was inspired by the design of Praat on the Web (Domínguez et al., 2016).
We gratefully acknowledge their contribution to web-based speech annotation platforms.
Praat++ is developed by the ACL Lab at the University of Texas at Arlington (UTA).
