Computational Memory Lab: PyParse

PyParse Behavioral Sound Parsing Program The PyParse program is for precise scoring and timing of participant responses during verbal tasks, especially Free Recall. It displays the recording of the participant's responses as a waveform on the screen, which is traversed by a cursor as the file is played back.

PyParse uses the wxPython GUI programming library to display the window and its components, and the SciPy library to filter the waveform before displaying.

Installation Instructions

Mac OS X

Download and run the installer for the latest MacPython 2.4 from pythonmac.org.
Download and run the installer appropriate for your system:
- PowerPC Mac
- Intel Mac

Other Systems

Step 1: General Dependencies

Install the following:

libsndfile - You may download it from here.
libsamplerate- you may download it from here.
swig - you may download it from here.
ALSA

On Ubuntu linux, these are provided by the packages:
libsndfile1 libsndfile1-dev libsamplerate0 libsamplerate0-dev swig libasound2 libasound2-dev

Step 2: Python Dependencies

Install wxPython, Numeric, and scipy as appropriate for your distribution.
On Ubuntu linux, these are provided by the packages:
python-wxtools python-numeric python-scipy
Download and unpack the PyParse tarball.
On Ubuntu linux: sudo python setup.py install.
Otherwise:
- Become root
- cd into the top pyparse directory
- type: python setup.py install

Sample Data

You can try out your installation with this handy sample data.

If you have any questions about these pages, please e-mail our webmaster.