Booth SU11706CM Manual media processing activities such as QC, segmentation of long-form video and VoD conversion have become an essential part of content preparation phase in the broadcast workflow today, says Amagi. However, these processes require a large number of specialist professionals, making it not only expensive but also time-consuming.
Amagi Tornado leverages machine learning models that can be trained for activities such as detection of black frames or color bars, create VoD assets from live stream in near real time, and automatically detect ads in incoming linear streams to facilitate ad insertion in the absence of ad markers, among others.
Amagi says its Tornado is a secure platform, hosted on a virtual private network within public cloud, with permission-based access safeguards. Tornado uses machine learning models that are trained using thousands of hours of content across genres, resulting in high-confidence output validated by video specialists.