SpeechView is a component of the CSLU Toolkit used for displaying, playing,
and editing speech waveforms. It can also display spectrograms and other
types of data related to speech waveforms such as pitch and energy contours,
neural net outputs, and phonetic labels. SpeechView can read previously
created waveform files; it can also record new waveforms and save them to
disk. The basic functionality provided by SpeechView is used in other
CSLU Toolkit components such as
BaldiSync and the
Rapid Application Developer.
The top-level functionality of SpeechView is accessed via a
main menu which permits the reading of
waveform files and recording of new waveforms, and also by
a
toolbar which allows for these waveform
manipulations as well as other common SpeechView functions. Underneath
the toolbar, the application display area consists of zero or more waveform
groups, one for each different
waveform currently loaded. A waveform group consists of all of the windows
connected to a given waveform; these windows can be classified into two major
sets of
window types:
-
those provided by default
with each waveform group, and
-
those optionally requested
by the application user.
SpeechView is
equipped with a
command-line interface
that allows the user to exercise more control over the appearance and
functionality of the application than is available with the default
application.
The example SpeechView application displayed above consists of the toolbar
with two different waveform groups, each delimited on top by a scrollbar. The
first waveform group includes the default windows plus a user-requested
spectrogram window. The second waveform group includes only the default
window types.
If you have not used SpeechView, all of these options may seem overwhelming
at first view. To help you get started, there is a short
tutorial.
For compatibility with previous versions of SpeechView, there is
a version with no toolbar and a set of sidebar menus.
