Colombia. After a careful evaluation of ENCO's enCaption4 automatic captioning system at four trade shows, Colombian TV channel Citytv opted to end the challenges of manual subtitling. The TV station serving the capital city region of Bogotá has increased speed and accuracy while reducing subtitle costs by 75%.
Citytv, owned by the media company Casa Editorial El Tiempo, licenses the Citytv brand from the Canadian communications company Rogers Media. The station, which broadcasts on UHF channel 21, follows a mandate that requires Colombian over-the-air television stations to provide subtitles on 75% of their programs between 6 a.m. and midnight. Citytv subtitles its daily news program, Citynoticias, along with other popular programs such as Arriba Bogotá and Bravissimo.
"We have closely monitored the development of enCaption4, and it has clearly reached the level of accuracy we require," said Walter Gomez Zuluaga, Engineering and Operations Manager, Citytv. "We now have a solution that fully automates the subtitling process, better serves our hearing impaired viewers, and ensures we meet all regulatory obligations."
The speed with which enCaption4 generates subtitles, especially compared to the manual "listen and talk" voice recognition system it replaced, was a key factor in Citytv's decision. Based on ENCO's proprietary auto-subtitling approach, enCaption4 combines machine learning with a neural network speech-to-text engine to deliver exceptional accuracy with extremely low latency.
Citytv also interconnects enCaption with its MediaCentral (iNews) newsroom (NRCS) computer system to familiarize you with specific words you will interpret in subtitles. Using the standard MOS protocol, enCaption4 can access news scripts and summaries from any NRCS and learn correct vocabulary, spelling, names, places, and other special content to be mentioned in local newscasts. It retains these terms in its ever-expanding internal dictionary, which becomes more valuable to the user over time.
Continuous product development has also resulted in enCaption4's ability to distinguish between various people who speak. This new multi-speaker distinction functionality leverages artificial intelligence to detect changes between speakers, even within a single mixed power. enCaption4 today supports more than 30 languages with accurate spelling, capitalization and punctuation to strengthen the quality of subtitles.
