InBroadcast - January 2023 - 43

www.inbroadcast.com | Vol: 13 - Issue 1 | January 2023
43
InMedia
Closed Captioning in global
markets: supporting SDI
and IP workflows
K
eeping a close eye on
industry
shifts
continually
and
subtitles
everchanging
needs, VoiceInteraction
is
working on
expanding its coverage to new
production and
distribution
workflows. As the Broadcast
Industry accelerates its move to
IP-based production workflows,
Audimus.Media can now operate
as an ST 2110-40 captioning
stream generator, with native
SMPTE 2110 support added. To
reduce the time-to-market and
quickly carry the software from
SDI to ST 2110, a partnership with
Deltacast was put in place via their
virtual card, thus ensuring the most
agile solution possible.
For new IP-based workflows
Audimus.Media
is
deployment in any
IP
ready
for
broadcast
infrastructure through Deltacast's
IP Virtual Card. This new video and
multimedia programming interface
allows the use of 3rd-party computer
network cards to capture and stream
out video. The fact that this solution
can be installed on any machine,
regardless of the NIC card, was a
key decision factor for the company.
Furthermore, the VideoMasterIP SDK
provided by
Deltacast makes the
integration process seamless: the
SDK connects the application to the
IP Virtual Card through an easy-touse
API. The IP Virtual Card brings
in support for video transport as per
ST2110-20, PCM audio transport as
per ST2110-30, and ancillary data
transport as per ST2110-40.
VoiceInteraction develops state-ofthe-art
Speech Processing solutions,
producing automatic Closed Captions
in real-time for live broadcasts and
other media platforms. With high
into any
Non-Linear
Video Editor with automatic
clip markings. Responding to
the challenges every client face
remains of utmost importance,
as
regulatory
demands
rise
globally, extending the need to
provide closed captioning to live
broadcasts all over the world.
Beyond signal input
infrastructure - VoiceInteraction
proprietary technology
Aside from expanding
coverage to new production
and
accuracy and low latency, Audimus.
Media is VoiceInteraction's flagship
software for the Broadcast Industry.
This
accurate, reliable, and cost-effective
captioning - now with live translation
on-premises; source
playlist
creation
CTA-708 captions
AI-driven platform provides
video signal
encoding to enable clip export
with synchronized captions to VOD
platforms; HLS
with WebVTT files to allow stream
accessibility in any language, and
much more.
Coping with market specificities
Audimus.Media has
traditionally
external
associated
closed
been
with an
caption encoder/
Teletext inserter device to add the
automatic subtitles to the SDI video
signal. With hundreds
of
software
clients
globally, the R&D team have been
developing new updates not only
to accommodate recent
developments
but
to the industry,
also streamlining hardware
requisites in order to provide the
smoothest experience possible to
Broadcasters who are looking to
maintain their current structure. For
markets with restricted
budgets,
can now be
encoded into SDI signals while still
offering a caption monitoring output,
depending on the card used. Given
the ongoing transition to IP-based
production workflows, Audimus.
Media can now operate as an ST
2110-40 captioning stream generator,
with native SMPTE 2110 support
added.
VoiceInteraction has
been
expanding the native formats that
can be produced and sent as a
contribution to MPEG-TS multiplexers
or muxed by Audimus.Media into a
MPEG-TS stream: DVB-Teletext, DVBSubtitling,
ARIB B24, ST 2038:202,
and ETSI EN 301 775.
New supported transport protocols
were also added: Secure Reliable
Protocol (SRT) and Reliable Internet
Stream Transport (RIST) can be used
as input or output.
One of the longstanding challenges
for our customers has been the
distribution of live captions in their
VOD platforms, with seamless audio
and
latest
video synchronization. Our
update combines
product
encoded video recording with an
embedded
editor, exporting the
distribution workflows,
the team continues to develop its
core speech processing technology.
VoiceInteraction platforms rely on
our proprietary
Automatic Speech
Recognition (ASR) engine, AUDIMUS.
Constantly
for highly
evolving, it
allows
customizable speech
processing tasks in real-time, as well
as post-recording workflows in over
30 languages.
Adding to an advanced
decoding strategy
new
that
allows for
increased accuracy in unprepared
speech, speech translation is now
produced for live sources, enabling
multiple subtitle languages
source stream. The underlying
proprietary
and
speaker identification
language identification
modules were overhauled, for new
classifications produced with lower
latency and higher accuracy.
VoiceInteraction creates and refines
different language models based on
classical
methodologies, Machine
Learning and Deep Neural Networks.
These are updated daily
through
an innovative deployment pipeline,
bringing a clear added advantage to
our clients.
https://www.voiceinteraction.ai
per
http://www.inbroadcast.com http://bit.ly/3X7mQXI http://bit.ly/3X7mQXI https://www.voiceinteraction.ai http://bit.ly/3X7mQXI
InBroadcast - January 2023

Table of Contents for the Digital Edition of InBroadcast - January 2023
