A System That Solves the Challenges of Conventional Real-Time Captioning
Overview
“Solo” and “Accurate” Live Captioning.
Equipped with an AI speech recognition engine and an intuitive UI, dramatically reducing the cost of live broadcast captioning operations.
“Mojipa,” which has redefined “real-time caption broadcasting,” enables “solo caption production” through intuitive and simple touch panel operation.
※ “Mojipa” is a registered trademark or trademark of TBS Television, Inc.
System Overview
Features
■ AI Speech Recognition Engine
“Mojipa” uses an AI speech recognition engine to analyze audio input and automatically generate captions in real time.
Using a speech recognition engine compatible with both on-premises and cloud environments, it reduces the burden on operators.
Actual Broadcast Screen
Input Screen
■ Labor Savings and Cost Reduction in Caption Generation
With conventional methods, a specialist with specific skills was required. By using “Mojipa,” operators simply verify and fine-tune the captions displayed on screen, making caption generation straightforward.
The collaboration between AI and humans enables fast and accurate caption generation.
How It Works
■ Simple and Easy Operation
Equipped with a touch panel UI to support real-time caption generation where speed is essential.
Through intuitive operation — simply touching characters on the display — users can correct and broadcast auto-generated captions.
“Mojipa” can be operated with a single finger on the touch panel.
Intuitive operations such as tap, flick, and swipe enable “solo caption production.”
Additional Features
Delay Display
Displays the delay time between the actual broadcast and caption generation.
Auto TAKE Function
Set the broadcast interval to enable automatic TAKE.
Template Text Insertion
Insert pre-registered phrases with ease.
User Dictionary Registration
You can use a user dictionary for displaying prediction candidates in Mojipa, a speech recognition engine dictionary,
and the PC’s IME dictionary for text input.
