Enabling Speech-To-Text YouTube Videos Enhances Business Potential

Catalog's Logo Logo of Moji-Moji TV
On Wednesday, Tokyo-based tech start-up in media rep service and speech recognition technology development Catalog just released the closed alpha version of the brand new service that enables speech-recognize what is spoken in a movie on the video sharing services such as YouTube and Nico Nico Douga[J].   In this release, the first 1,000 applicants can get access to the service to evaluate its functionality and usability.
The service is named “Moji-Moji TV[J]“.   It captures audio in a movie and to result it as a text output, which can be used for subtitles in a movie as well as for describing and tagging the movie itself.   Moji-Moji TV also has a remarkable feature that collects people’s names, newly-coined words and slang expressions from the Internet and add them to the dictionary of the service’s speech-to-text engine.
Referring to comments to be given by the alpha test users, the company plans to improve and enhance the service for its official release which is scheduled in several months.
Incidentally, Google released a gadget called Election Video Search last summer, which allows you to search through a bunch of YouTube videos for what was spoken, using Google’s speech-to-text technology (only for American-English speaking).

Leave a comment

Your email address will not be published.