What is RACU?
RACU stands for “reconnaissance automatique avec corrections ultérieures” (automatic recognition with retrospective corrections). STXT provides the RACU workflow as an accessibility tool for the MV project.
How does it work?
The process starts with automatic speech recognition of the uploaded videos. Afterward, the text is machine translated and automatically segmented into subtitles. The multilingual subtitles are made available as SRT or VTT files for download.
What are the callenges?
In order to access the RACU process, it is necessary to log in to MediaHub – a central workflow system developed by STXT in past projects.
In addition, the system was developed for broadcasters and is too complex since it requires extensive training and is not intuitive to use.
What is the solution?
The RACU process will be integrated into a new MediaVerse system specially designed for semi-professional users’ needs (prosumers). This will make the whole process is transparent to the user when uploading the content.
The RACU system in the MV will consist of 3 components: subtitle workflow, translation workflow, and subtitle editor (see fig. 1)
The subtitle & translation workflow will be provided as a backend service addressable via the MV API. This way, a user receives the corresponding subtitle files in several languages for a video. These automatic processes run transparently in the background for the user.
The subtitle editor will be developed within the MV project. It can be used independently as a stand-alone service and allows an intuitive edition of subtitle files. It is also integrated into the MV ecosystem and accessible via the DAM. It will also provide a subtitling experience specially designed to interact with automatic recognition and machine translation.