The speecheditor was built in the VisLab at UC Berkeley by Steve Rubin in collaboration with Maneesh Agrawala and Floraine Berthouzoz at Berkeley, and Gautham Mysore and Wilmot Li at Adobe Research. The interface is described in detail in our UIST 2013 paper, "Content-based Tools for Editing Audio Stories."
The speecheditor relies on radiotool, a python library for manipulating audio. Radiotool was also developed by Steve Rubin in the VisLab.
Vagrant requirement was added to the codebase to make it easier to get up and running with the speecheditor. You can also install the requirements on your local system (Mac or Linux) but the
Vagrant method is much easier to debug.
For more information, contact Steve Rubin <srubin at cs.berkeley.edu> and Maneesh Agrawala <maneesh at cs.berkeley.edu>.