LINGUIST List 21.3735

Tue Sep 21 2010

Software: Prosody Research Tools Available

Editor for this issue: Dayn Schulert <daynlinguistlist.org>


        1.    Yi Xu, Prosody Research Tools Available

Message 1: Prosody Research Tools Available
Date: 15-Sep-2010
From: Yi Xu <yi.xuucl.ac.uk>
Subject: Prosody Research Tools Available
E-mail this message to a friend

We are pleased to announce the availability of two Praat script tools forconducting large-scale prosody research: PENTAtrainer and ProsodyPro, bothfreely accessible at http://www.phon.ucl.ac.uk/home/yi/tools.html.

1. PENTAtrainer (http://www.phon.ucl.ac.uk/home/yi/PENTAtrainer/)

An interactive script for automatic extraction of pitch targets for prosodysynthesis. It is based on the qTA implementation of the PENTA model, and itallows users to:

- Automatically extract pitch target parameters (slope, height, strength)- Resynthesize F0 contours based on the extracted parameters- Specify target location and restrict direction of target slope- Exhaustively process all wave files in a folder- Perform the same f0 analyses as ProsodyPro (See below)- Automatically collect extracted parameters from all sounds in a folderand save them in ensemble files for ease of further analysis and processing

2. ProsodyPro (http://www.phon.ucl.ac.uk/home/yi/ProsodyPro/)

An interactive script for large-scale systematic prosody analysis. This isan updated version of its predecessor TimeNormalizeF0. Among its manyfeature are the following:

- Semi-automatic extraction of F0 contours, allowing users to improve theaccuracy of F0 tracking by rectifying vocal pulse marking- Exhaustively process all wave files in a folder- Automatically save time-normalized F0 and intensity contours for selectedintervals- Automatically save continuous F0 velocity contours for selected intervals- Automatically save many prosodic measurements, including: maxf0, minf0,f0range, meanf0, mean intensity, duration, max velocity, final velocity,final f0 and mean intensity- Automatically collect extracted contours and measurements from all soundsin a folder and save them in ensemble files for ease of further analysisand processing

Explanations and instructions can be found at the web sites. For moreinformation and bug reports, please contact yi.xuucl.ac.uk

Linguistic Field(s): Cognitive Science                             Computational Linguistics                             Phonetics                             Phonology                             Text/Corpus Linguistics

Page Updated: 21-Sep-2010