VOD (Video on Demand) Applications
As broadcast moves from solely using terrestrial or satellite transmission, to also being carried over the internet, and consumed on mobile devices, there is a major paradigm shift. Traditional broadcasts worked with a schedule, but internet-based broadcasting is largely an on demand service – viewers choose what they wish to watch, and when they wish to watch it.
File preparation for VOD has some characteristics that differ from regular broadcast, dictated by the different playout mechanisms employed. A common requirement is for files to need to contain 16 channels of audio, even when only a single stereo is being provided. Frequently, that stereo has to be on channels other than 1 & 2 in the file. This leads to a common workflow as used by a number of our customers, whereby the original content is created as a two channel MXF, and this has to be converted to a sixteen channel MXF, and the audio on 1 & 2 moved to different channels. The following workflow is typical –
How is this normally done?
Although a very mundane and tedious process, a number of companies would do the conversion from two channel to 16 channel MXF in an Edit suite. Whilst this may be a fine approach for one or two assets, it is a very poor use of an expensive Edit suite and a skilled editor, who is far more productive when using their time for true creative work.
Whilst Engine also includes tools to let you select and process individual files, for high volumes Engine also includes as standard a watch (drop) folder facility. Once configured, you simply drop your two channel MXF files into the watch folder, then get newly created files placed where required, including optional XML or PDF reports if required.
Alternatively, you can take advantage of the flexible API available with Engine, and fully integrate these functions into your MAM tools. Engine processing runs as a Service or Daemon for maximum robustness and 24/7 reliability.
Engine’s processing speed for this kind of application is variable based upon factors such as the speed of the storage and the size of the video – HD files take longer to process simply because much more video data has to be copied in to the new file. However, normal operation within a typical Broadcast facility produces processing speeds VERY roughly around 1.5 times faster than real time, per Engine processor. Up to eight Engine processors (ESPs) may be purchased, suitable for very high volume applications. Each ESP processes one file at a time, so eight ESPs let you process around 12 hours of content per hour, and of course this is an entirely unattended operation.
What about audio levels?
Worldwide broadcasters now are reasonably standardised on setting Program Loudness to -23 LUFS or -24 LKFS. This is ideal for delivery to the sitting room, but for mobile consumption, where consumers could be using basic headphones, and in noisy environments, some broadcasters wish to target a higher level, and -16 LUFS is commonly used. Using Engine, this can be easily achieved at the same time as the first requirement.
In all cases, the output from Engine will be an MXF file, with the original video and general metadata fully preserved.
Audio Description (AD)
Another aspect of VOD is the need to supply AD commentary with the content. In terrestrial broadcast, this is supplied as an additional soundtrack, and is mixed in the users hardware. Mobile devices do not include hardware for mixing the commentary track with the main soundtrack so this has to be premixed. Engine has a module for this function, which can of course be incorporated into the above workflows if required. Read about AD here