Research

About Fluency Profiling System: Speech is a dynamic, embodied cognitive system requiring the coordination of many cognitive processes unfolding over time. The temporal characteristics of speech can be captured by examining distributions of the durations of measurable speech components, namely speech and pause duration. However, several barriers prevent easy analysis of pause durations: The first is the issue of finding the minimum duration that counts as a 'real' pause segment. The second is the distribution threshold problem, which is the problem of appropriately classifying the pause segments as either short or long while minimizing the classification error rate. The Fluency Profiling System is a fully automated system for determining the speech-pause transitions and estimating the temporal parameters of those distributions.

Details about the FPS can be found here:

Little, D. R., Oehmen, R., Dunn, J., Hird, K. & Kirsner, K. (2011). Fluency Profiling System: Fluency Profiling System: An automated system for analyzing the temporal properties of speech. Manuscript in preparation. The University of Melbourne & The University of Western Australia.[pdf]

Technical information
To install the Fluency Profiling System, follow these instructions:
1) Download the latest version of the FPS below (rar file)
2) Unrar the compressed file. (If you do not have winrar installed, you can download a free trial here: http://www.rarlab.com/download.htm)
3) If you do not have the Matlab Component Runtime installed on your system, download the MCR installer below and install.
4) Follow the instructions in the "FPS_Installation_Instructions.pdf" to run the FPS.

Download FPS v6.00

Download Matlab Component Runtime Installer

Developed by:
Prof. Kim Kirsner (The University of Western Australia)
Prof. Kathryn Hird (University of Notre Dame & The University of Western Australia)
Prof. John Dunn (The University of Adelaide)
Dr. Daniel Little (The University of Melbourne)
Raoul Oehmen (The University of Western Australia)