Hints for recording a lecture video
This section is not particularly focussed on Blender.LectureEdit, but it is a compilation of general hints about recording a lecture video. You can see this as a checklist, that you go through when preparing and conducting a recording session. However, this list does not claim to be complete.
Audio
The audio quality is often overlooked when recording a video, even though it might be the most important issue during a recording session. Watching a lecture video with distracting background noise or hardly intelligible speech makes it hard to learn about the lecture’s subject.
Do not underestimate the room’s reverberation
The human hearing is very good at adapting to different situations. Because of this, we often perceive rooms as far less reverberant, than they actually are. However, if you listen to a lecture, that was recorded in a reverberant room, the adaptation does not work, because you are not in that particular room at the time of listening. That way, the reverberation in the recording becomes very apparent and often annoying.
Therefore, it is recommended to do a short test recording and listen to that recording in a different room to assess the audio quality.
If there is too much reverberation in the room, you have three options:
Record the lecture in a different room
Treat the room acoustics by adding absorbers
Make sure, that the microphone picks up mostly the direct sound from the speaker and not the room’s reverberations
Often, only the microphone can be changed in a recording session. There are generally two strategies, that you can apply:
Move the microphone closer to the speaker. This is the main idea behind lavalier and headset microphones. Both are great for recording lectures. If you place a regular microphone in front of the speaker, this works just as well, but it is much more apparent in the video. However, as long as you place the microphone and the stand in an aesthetically pleasing way, this does not spoil the video.
Use a highly directed microphone, that only picks up sound from the direction of the speaker.
When using a single microphone, you are probably looking for microphones with a super-/hyper-cardioid directivity pattern. Shotgun microphones are even better in this regard, although the good ones are relatively expensive.
Another way to influence the directivity of the sound pick up is to use multiple microphones and combine their signal to attenuate sound from unwanted directions. This is, what mobile phones and laptop computers do for their telephony functionality. There are also some conference microphones, that also use this technique. All of these devices are cheap or already available and they are optimized for speech recordings. So this is an inexpensive way of achieving a good speech intelligibility. However, the studio microphones from the aforementioned category may have a fuller, more natural sound.
Check for undesired noises
Especially with lavalier microphones, that are attached to the speaker’s clothes, it often happens, that the clothes or the speaker’s hair rub against the microphone. This causes a loud rumbling sound, that completely masks the lecturer’s speech. The same issue can occur with headset microphones, when the speaker has long hair and/or a beard.
Another issue can be, that the microphone picks up undesired noises from the speaker like chewing, smacking or breathing. Headset microphones seem to be particularly prone to this issue.
And if the microphone is close to the speaker’s mouth, plosive consonants like P and T can cause the speaker to blow in the microphone. This causes loud and distorted noises.
So, if you use such a set up, it is recommended to do a test recording, in which you try to provoke the aforementioned issues. Most issues can be mitigated by positioning the microphone slightly differently. You can use pop filters against the issue with the plosive consonants, however, they will likely look awkward in the video recording.
When recording outside, wind noises become a severe issue, even with low wind speeds. Most (semi-)professional microphones are shipped with a piece of foam to mitigate this issue. You can even craft a simple wind screen out of a sponge. For higher wind speeds, you need a wind screen, that is colloquially referred to as “dead cat”.
Do not record in noisy environments
Background noise is distracting and will severely impair the speech intelligibility. Therefore, you should avoid recording in noisy environments. If you have to be in a noisy environment, e.g. for a live demonstration, confine the noisy parts of the recording to only those parts, where you have to be in that environment and record the rest in a quiet place.
If you are recording at home, make sure, that noisy devices like the dishwasher are switched off and no-one flushes the toilet, while you are recording. Also, close your windows to block out noises from the road.
You can reduce the problems with background noise by following the aforementioned hints about microphone placement and using a directed microphone.
Always have an audio track
When creating a lecture video, you have to synchronize multiple video files. And the easiest way to do that is to align their audio tracks. So make sure, that you always record an audio track alongside each of your videos. Except for the speaker video, the audio does not have to have great quality, since it is only used for synchronization.
Video
It is advised to spend some effort to ensure a good video quality. This is especially important, if you are doing a green screen recording, because a good video quality greatly increases the reliability of the green screen processing.
Don’t wear green
Anything green or green-ish (that often includes yellow) can be confused by the green screen processing with the background color. So it is best to pick dark clothes, preferably in red or purple, because that facilitates the distiction of the speaker from the light green background of the green screen.
If it is not possible to avoid something green on the speaker (e.g. because she/he has dyed their hair in such a color), you can use a different background color. A common alternative is a blue screen. Some colors may be problematic though, like white, which may also occur in the foreground, when light is reflected by the speaker’s glasses or jewelry. Especially avoid colors, that are close to the speaker’s skin color.
Avoid fine patterns
A camera has a limited resolution, that is not capable of capturing fine patterns in the speaker’s clothing. For example, thin stripes on the speaker’s shirt will lead to a confusing Moiré pattern.
Proper illumination
Both the speaker and the background have to be properly illuminated. More light is generally a good thing, because it increases the quality of the camera image. In addition to that, there are some spots, where special care is advised:
Avoid shadows in the speaker’s face. For example, if you illuminate your recording only with the ceiling light of the room, it is likely, that the speaker has shadows below their eyes. This way, the speaker will look tired, sick or hung over. Therefore it is recommended to have an additional light source, that illuminates the speaker from slightly below.
Avoid shadows from your equipment. It is usually desirable to have some distance between the speaker and the light source. This softens the shadows and reduces issues with reflections. However, if your light source is behind your camera, you will have a somewhat prominent shadow in your recording. In such a case, you can move the light sources more to the front and/or use additional light sources to illuminate the otherwise shaded spots.
Avoid shadows on the green screen. If the speaker casts a shadow on the green screen, these areas will have a different color than the illuminated areas of the green screen. Such contrasts can impair the green screen processing. Therefore it is recommended to have some distance between the speaker and the green screen and set up a light source behind the speaker to avoid such shadows. If you put the light source too close to the green screen, you will have large shadows from wrinkles in the green screen, so make sure, that the green screen is as smooth as possible.
Don’t crop too much during the recording
Let’s assume, you only want to show the speaker down to their chest in the final video. You should not do this cropping during the recording by only recording the speaker’s face and chest. If you record the speaker down to the hips, you have more liberties during the processing of the video to position the speaker exactly where you want them to be. This is especially important, if the video is recorded in multiple takes, so you have to harmonize the cropping from multiple recordings.