The
Voice Writing Method of Court
Reporting
Components of a Voice-Based CAT System
Basic System
The base components of today's voice writing CAT systems are a
stenomask, a laptop or desktop personal computer, entry-level CAT
system and a speech recognition engine (SRE), which can be purchased
for well under $7,500. This suite of equipment is generally used
by a reporter who would use a typist to produce the transcript from
both the English file and the reporter's voice file. The voice
writer's voice file is the direct equivalent of the stenotypist's paper
notes; these are now viewed as backup records, archived as required by
law.
Realtime
Realtime voice writers generally purchase the most powerful laptop
computers available, the "professional" versions of CAT software and
speech recognition software from IBM or Scansoft, an external USB,
DSP-based sound card, and remote display hardware/software, such as
LiveNote. This suite of equipment is generally used by a reporter
who produces realtime output, either for herself, for multicast
transmission to other screens in the courtroom, or other external
destinations.
The word "realtime" requires a bit of defining here. IBM's speech
recognition software is realtime by nature, meaning that English words
are identified and sent to the screen usually within milliseconds of
being spoken by the reporter, as IBM's software process each word
separately. ScanSoft's Dragon Naturally Speaking software "batch
processes" whole phrases, sending them through an AI engine so as to
increase the accuracy percentage. This usually results in a delay
of a few seconds for longer phrases to appear on a screen. The
savvy reporter inserts phrase-ending punctuation, such as commas or
periods, to force a quicker phrase analysis for faster display.
Thus, a voice reporter using a speech recognition engine is always in
realtime mode, from a purely technological perspective, despite the
traditional meaning of where the reporter directs his or her English
output.
Captioning and CART
Captioners and CART (communications access realtime transcription)
providers usually use a high-end suite of equipment, but they add a
Line 21 television captioning decoder. Captioners also connect
their computers to a TV station via modem or other Internet
connection. Voice writer captioners produce both traditional
captioning, and also streaming text for live Internet webcasts.
Pre- and Post-Production
Voice writers use the entire range of post-production software used by
their stenotypist cousins, including Amicus
and other forms of ASCII conversion, Summation, eTranscript, et
al. Between stenotype and voice writers, the only aspect of the
profession which differs is
the method of taking down testimony; all pre- and post-production
aspects of the profession remain the same.