|
The
Annotated Text Processor (ATP)
(click
to enlarge)
ATP has a media player interface that
allows the user to play sound and video files and to
link them directly with related text documents. The
underlying media player control supports WAV, MPG, AVI,
and MOV formats.
The "Select" button allows
the user to browse and open the desired media file while
the "Close" button flushes the media buffer
and closes the file. The navigation bar
sits just below the path name and provides "Play,"
"Pause," and "Stop" buttons. The
length of the media file—in this case a WAV sound
file—is reported in the "Length" box
when it is opened. All durations and positions are reported
in milliseconds (thousandths of a second). Thus the
sound file in the example is 457.643 seconds (00:07:37.643
h:m:s) in length. In practice, the media player has
been tested successfully with sound files of up to an
hour in length. Whenever the user clicks the Play, Pause,
or Stop button, the current position is reported in
the "Position" box. (Both the Length and Position
boxes are read only.)
While the sound is playing the linguist
can click the "Mark" button
to learn the position of a segment boundary or interesting
phonetic feature without stopping or pausing the sound.
The "Play on Mark" button will play a predefined
segment of the sound file—three seconds in this
case—centered on the Mark itself. If the Mark
does record a segment boundary, the linguist can use
other buttons can transfer the Mark to the Start
and End boxes to define the segment
exactly. The Play button always plays from the Start
position to the End position. (Initially Start is zero
and End equals Length.) Start and End can also be incremented
or decremented by predefined amounts, or changed by
hand. The Start and End positions of each segment can
be saved with the text data and reloaded from it as
long as the data model provides storage locations. This
is how the linguist actively links sound and text data.
The media player serves purposes at
three stages of linguistic analysis:
- The general goal is to produce a
transcription of the audio or video recording but
at the earliest stage the linguist must discover
how to divide the untranscribed recording
into phrases, sentences, and other appropriate segments.
The linguist can play and replay the media resource
and explore it at will, and decide on tentative segment
positions, and create a very tentative text structure
with minimal contents, all of which can be easily
revised on closer examination.
- Once the recording has been marked
initially, the detailed transcription
can proceed in earnest. The linguist listens carefully
to each marked segment and fills in and completes
the text, adjusting and finalizing the associated
media-file positions as he or she proceeds. On completion,
the linguist has both an accurate text transcription
and a fully analyzed sound or video resource which
are linked to one another in detail.
- Once the transciption of this text
is finished, and other texts are also transcibed and
linked to recordings, the media player controls facilitate
detailed and extensive searches of texts and
recordings in support of further comparative
or phonetic work.
In principle, the media player can
be extended to directly handle DAT, cassette player,
and CD audio, or any device that supplies a Media Control
Interface (MCI) driver. Such an extensions would only
require the addition of tools to handle tracks and relevant
hardware switches.
Back
to Annotated Text Processor
|
