SubtitleNEXT powerfully accelerates Live Subtitling with Google ASR integration

Automatic Subtitles can be Created directly in Live news events with SubtitleNEXT

We are happy to unveil that SubtitleNEXT’s enhanced Live Subtitling feature now includes the integration of Google’s automatic Speech-to-Text engine. This notable new development in SubtitleNEXT extensively sets it apart by further enhancing and substantially speeding up the entire Live Subtitling process by allowing videos to be translated live in real time at a stunning pace.

The new Live Subtitling function in SubtitleNEXT allows for immediate real-time post-editing of live recognised text, enabling a two-stage dictation workflow. It also adds support for user configurable latency of recognised text, and when combined with real-time post-editing features, the system facilitates high quality subtitles and captions with low latency.

SubtitleNEXT’s new capability enables automatic subtitles to be created directly from the voice of a speaker with the added benefit of now being able to further refine Google’s input to ensure high quality optimisation. This ability significantly accelerates the entire process to streamline live subtitles and captions for live television shows and events for live broadcast.

The SubtitleNEXT platform supports single and multichannel speech recognition provided by the Google Speech-to-Text API. The Google Speech-to-Text API is useful in separating several speakers and improves recognition in the case of simultaneous speech by multiple speakers, of which up to 8 speakers can be supported at a given time.

Industry professionals can use SubtitleNEXT in various workflow scenarios that include automatic or unattended subtitle and caption generation, manual operator-assisted with optional real-time post-editing, as well as semi-automatic combinations of these workflows.

Regarding sound, sources can operate within any audio input within SubtitleNEXT, ranging from microphones, through to analogue, digital, SDI embedded technologies, network streaming and pre-recorded audio from either video or audio files.

Profuz Digital’s CTO Kamen Ferdinandov and creator of the SubtitleNEXT software system notes, “SubtitleNEXT’s software has inbuilt settings in place to handle text speed and quality delivered from Google’s automatic speech recognition (ASR) engine. These significant Live Subtitling capabilities within SubtitleNEXT’s toolsets now provides subtitlers with the additional choice of evaluating and testing the most suitable and efficient results for live content. Subtitling is a crucial requirement to help facilitate accurate communication across the world today, and our team is continuously looking at ways of improving how users can work more efficiently and how our toolsets can benefit them and further enhance their skills.”

CEO of PBT EU Ivanka Vassileva adds, “SubtitleNEXT is a popular subtitling system that has already been adopted by many media, production, post facilities and academic institutions. We are enormously proud to provide high quality accessibility services that can be made attainable for larger audiences. This remarkable new development in SubtitleNEXT makes it a “best-fit” solution for subtitling preparation and for the airing of news shows for traditional broadcasters as well as for all types of accessibility services that include live events, including educational sessions, YouTube live events and much more.

The Live Subtitling function in SubtitleNEXT significantly speeds up the subtitling of news emissions and live shows in television and we have a few customers currently looking at using this specific workflow already with plans to extend it further and apply it to more channels which is exciting news, to follow soon.”

Watch the SubtitleNEXT Live Subtitling video here for further insight