[Asia Economy Reporter Eunmo Koo] SK Telecom and JTBC Studio have joined hands to improve viewing accessibility for the hearing impaired and to promote the overseas expansion of K-content.


SK Telecom and JTBC Studio Jointly Develop 'AI Video Subtitle Technology' View original image

On the 27th, SK Telecom announced that it will jointly develop a ‘real-time artificial intelligence (AI) automatic subtitle service’ technology with JTBC Studio that provides subtitle services for all programs and media. The real-time AI automatic subtitle service is a service where AI recognizes dialogue and other elements within video content to provide Korean subtitles.


JTBC Studio, which oversees JTBC’s content production and distribution, will provide video content for AI deep learning and broadcasting technology consulting, while SK Telecom will develop core technologies optimized for media, such as speech recognition and subtitle generation, to be used together.


Currently, major broadcasters provide subtitle services by having humans listen to the conversations in video content and transcribe them into subtitles. This has led to complaints about difficulties in proper viewing due to a delay of about 4 to 5 seconds and occasional omission of some dialogues. Additionally, this service has not been provided for on-demand video (VOD) and over-the-top (OTT) services outside of live broadcasts.


From the perspective of overseas expansion of domestic content, the technology includes automatic subtitle file generation and synchronization for not only scripted dramas but also educational and entertainment content, which can reduce preparation time and costs for exports. Furthermore, it is expected to play a positive role in spreading K-content by saving time and costs required for translation into the languages of export countries.


Currently, Google’s speech recognition is a representative subtitle generation technology, but it is difficult to use in noisy broadcasts and requires additional learning and development tailored to broadcasting content. In particular, for broadcasting content, dialects, background music, and ambient noise that interfere with speaker voice recognition lower the accuracy of subtitle extraction, necessitating specialized AI speech recognition and noise reduction technologies to address these issues.


SK Telecom and JTBC Studio Jointly Develop 'AI Video Subtitle Technology' View original image

SK Telecom utilized AI noise reduction technology developed in the first half of this year to remove noises such as wind, bird sounds, and car noises recorded during filming at the ‘SK Telecom Open Golf Tournament’ in June, vividly including conversations between players and caddies and player interviews in the live broadcast.


Through this collaboration, the two companies plan to expand the real-time subtitle service currently provided on major broadcasting channels to more channels and various media.


Park Changseong, Head of JTBC Studio’s Content Business Division, said, “It is meaningful that media technology development for improving viewer welfare can accelerate by utilizing JTBC content,” adding, “We will strive through close cooperation to contribute to the development of the domestic content industry.”


Lee Jaegwang, Head of SK Telecom’s Media Business Support Group, said, “Through the partnership with JTBC Studio, we have received deep learning content and broadcasting technology consulting, enabling us to solve the biggest challenges in AI media technology development,” and added, “We hope this media technology will contribute to viewer welfare and the spread of Hallyu through K-content.”





This content was produced with the assistance of AI translation services.

© The Asia Business Daily(www.asiae.co.kr). All rights reserved.

Today’s Briefing