A lot of churches start off their live-streaming ministry by using a smart phone or tablet. These are fine proof-of-concept way to get into live-streaming, but soon, you'll see the limitations.
Among these are the fact that words that are shown on screens to the in-person congregation, don't show online. The first solution that comes to mind is to pan over to the screen or compose (an often awkward) shot that includes both the primary subject and the screen.
While you can improve this way of doing things with a teaching TV for the sermon, it's more limited than you might want. To make the shot work, the pastor has to be close to the TV and the text has to be large enough to be legible. It's better to key text on top of the camera feed.
Let's look at the different ways to do it, and the advantages and disadvantages of each.
Using worship software
One of the first ways churches try to do this is by using your church's current worship software. MediaShout, EasyWorship, ProPresenter, and many of the other options, have a "live video" feature. The idea is that you capture video and have the software put text on top of it, live. Live is relative, though. Even with newer computers, there is some processing time, so you'll typically see around half second of latency or more.
If you're doing this for the online congregation, and not the in-person one, a little latency is no big deal. Just delay the audio to match the video and all's well.
There is a downside of using this method. Once the pastor knows you can put text on live video for the live-stream, you're likely to be asked to show it in your sanctuary, at least occasionally. A half second doesn't sound like a lot of time, but it can be the difference between highly engaging video and distraction-filled video that looks like it was dubbed over from another language.
A half second doesn't sound like a lot of time, but it can be the difference between highly engaging video and distraction-filled video that looks like it was dubbed over from another language.
Even if you don't run into that challenge, there are others. To do this correctly, you'll want a totally separate computer and either an additional license or a site license of the worship software you're using. Then it's simply a matter of capturing the video from the worship software and encoding it for the live-stream, using either a software or hardware encoder.
Using software switchers
If your church has moved beyond smartphone live streaming, you might be using a software encoder like vMix, Wirecast, or OBS (Open Broadcaster Software) to capture video and encode it for your live-streaming host, already.
If that's your situation, consider using that software to key out the background of the output from your worship software to put text on top of the video you've captured into the encoding computer.
Here, we've got some options.
While normally, you wouldn't want to run worship software and encoder software on the same machine (because both tasks can be processor-intensive, and therefore less stable when done on one machine), it's possible to do so and capture the output of the worship software and key it over the video captured by the encoding software.
To do so, send the right type of background. A black (or white) background can be removed as a luminance (or luma) key. Ideally, you'd remove only absolute black, but leave dark grays and other colors to maintain visual separation from the video in the background. Sometimes, this isn't possible, but it's what you should try to do, making sure to use an outline on the text for the purposes of this separation.
If your software doesn't allow you to dial in the color you removed to only the darkest (or brightest) colors, you might consider chromakey instead.
If your software doesn't allow you to dial in the color you removed to only the darkest (or brightest) colors, you might consider chromakey instead.
Just as your local television station removes a green background behind their meteorologist (and comic book movies remove it from behind actors in their scenes), you can remove green (or blue) from behind your text.
Neither of these techniques allows for anything other than a binary, all-or-nothing, transparency. If you want that, for example so that you can show a gradient behind a lower third, you'll need to use an alpha channel.
If you're sending a transparent background either between networked computers, using NewTek's Network Digital Interface (NDI) or something like Syphon or Sprout on the same computer, you can send alpha (if your worship software allows you to). This happens in the background with an extra channel of video.
The downside of this method is that not all encoding software and not all worship software support it. So, while it might seem ideal, it might not be possible for your church without some changes in the tools you use.
Hardware switchers
The best way to key is with a hardware switcher (not an inexpensive video selector, but full video switcher). While they're generally more expensive than software, since they're purpose-built they tend to be more reliable and more flexible than software (for example, an OS update or virus isn't going to affect them, like it might software).
The best way to key is with a hardware switcher (not an inexpensive video selector, but full video switcher).
If you don't already have a switcher, shop carefully to make sure it includes, not only the number of inputs you need for cameras and other video sources, but for your worship software, too.
Also, don't assume that any switcher can key using all three methods. Entry-level switchers, especially, don't always.
While luma and chroma keying work similarly to the way that they do in software, a linear key, using alpha, is a little different (unless you're getting it via NDI, for example). You'll need two video signals (and therefore two video inputs) to do this method.
Think of it, if you're familiar with Photoshop, as a layer and its mask. The layer itself contains all the information. Let's imagine this is an image with white text, a black stroke (or outline) around the text (in this case with the edge feathered), and a black background. If the color of the outline and the background are identical, you wouldn't be able to see where one ends and the next begins. Now, apply a mask. In the mask, the text is still white, but the outline is also white, slowly transitioning to gray then black. The background remains black.
In this situation, Photoshop would interpret the white text and part of the outline in the mask as opaque. The gray feathered part of the outline is translucent. The black part of the outline and background are transparent.
Getting a double signal, perfectly in sync, from your computer and its worship software is often done with plug-ins (that might cost extra) and additional hardware.
Without both signals, called key and fill, you can't get translucency.
Getting a double signal, perfectly in sync, from your computer and its worship software is often done with plug-ins (that might cost extra) and additional hardware.
So, while a linear key with a hardware video switcher is the preferred method, it's also the most expensive.
Choosing the right keying method
Like buying a new computer, it's best to get the most flexible keying method you can. This does two things, it makes your system more future-proof AND more leader-proof. If your church spends money so you can key, it's hard to tell the pastor, "Sorry, but we can't do that gradient lower third you saw at a conference you went to because I was afraid to ask for the ability to do a linear key." That's not to say that it's appropriate for every church to spend thousands of dollars to get a better key.
Know the priorities of your church. Pristine video with an unfunded youth ministry isn't a good combination. Just make sure everyone knows the limitations of the video system you have, so there are no surprises and you limit distractions.
