Content area
Full Text
The Praat acoustics program (Boersma, 2001) is powerful freeware that is widely used by behavioral scientists working with digital sound. This article describes GSU Praat Tools, a script package that helps simplify and automate such work. The routines use Praat's scripting language to create new menus and commands within the existing interface, and can operate either on individual files or in batch mode. The new functions facilitate selecting, displaying, editing, filtering, and otherwise modifying sounds, quantifying acoustic features, and saving results in text-based data files. The package includes an installation script and user's manual, and is available free from psyvoso.googlepages.com/softwaredownload.
The digital revolution has provided many benefits for the behavioral sciences, not the least of which have been in audio technology. Advances in this area have, for instance, made it much easier to record, modify, analyze, and synthesize sounds, as well as to use sounds in experiments. Furthermore, whereas there were once only a few programs available for working with digital sound, now there are many. The software includes a wide range of commercial, shareware, and freeware programs, offering a diversity of capabilities and features. As is often the case, however, this software exhibits a general trade-off between ease of use on the one hand and versatility and power on the other (see zeeman.ehc.edu/envs/Hopp/ sound.html for listings of a number of acoustics programs and related resources).
At the simpler end of things are programs such as Spectrogram (Visualization Software LLC), which digitizes sounds, presents high-quality graphical representations of waveforms, individual spectra, and time-by-frequency spectrograms, allows cursor-based measurements, and is practically effortless to learn. Raven (Cornell Lab of Ornithology) occupies an intermediate level hi providing elegant graphical displays and straightforward measurement of sound features, but also demands some study and effort by the user. The built-in measures are limited, focusing on basic time, amplitude, and frequency characteristics. SpeechStation2 (Sensimetrics Corporation) is similar in requiring some learning while providing a useful but smallish set of measures. Useful functionalities such as pitch tracking and resonance modeling are included in this program; graphics are simple and relatively inflexible.
The more powerful and flexible programs are those that, for example, provide direct access to the sound data, a variety of analysis and synthesis algorithms, and scripting capabilities for...