Tutorial Videos and Notes

I have recorded some quick tutorial videos to help with various projects that I'm involved in. I've included them here in case others might find them useful.

Contents:

General Audio Configuration

MacOS Audio

General Configuration

I'm using SoundSource on my Mac. It isn't free - it was $29 the last time I checked - but in my opinion it is well worth the price. SoundSource allows you to easily set the sample rate on your mics and also allows you to add boost, EQ, and some other effects. It isn't a full software mixer, but my Mac's hardware is configured for long battery life, not high performance, so SoundSource is plenty for me :)

Disabling ambient noise reduction in macOS Catalina

It seems that Ambient Noise Reduction (ANR) is on by default in macOS Catalina, unless you happen to have a 4-channel microphone. Unfortunately, Apple also removed the checkbox from the Sound preferences that would allow you to turn it off. As of now (09Oct2020) this seems to be a fairly well-known issue, and Apple has not fixed it. The only solution I've seen that has been confirmed to work is downgrading to Mojave, which I seriously considered, but support for it will most likely be ending soon. However, I stumbled onto a workaround: enable voice control, and if you don't actually want to use it, disable all the commands. (Update: Apple re-enabled the switch in Big Sur, but as of this update, the recommendation seems to be to stay away from it if you're a musician. Check the current situation if you decide to upgrade - Google is your friend. I have recently bailed out of the Mac world and will probably not be providing any further updates on this.) This video shows you how.

Windows Audio

General Configuration


Steinberg has an article on their site titled "Windows 10: How to set-up and optimize a Digital Audio Workstation (DAW)". You may find this useful if you're recording or streaming, especially if you have any audio processing software involved. (I'm using Voicemeeter Potato.)


If you have a Dell, apparently Waves MaxxAudio has a reputation for randomly taking over the entire CPU. Disabling that helped, but the Realtek drivers still seem to be messing with my system. I went into BIOS and disabled internal audio. Dell Support Assist is also reported to cause problems. (Update: I never did completely eliminate the problem. That computer is running Linux now.)


If you need to investigate what your Windows computer is doing in more detail, you may find Sysinternals useful.

Automatic Gain Control in Windows 10

(Note that it has gotten more difficult to remove automatic gain control and/or noise reduction in more recent versions of Windows and/or newer drivers. These functions may be bundled into a single "effects" setting which may also include other things such as volume boost. Make sure you test everything before you need it - don't assume that you know exactly what it did.)

Greek Orthodox Services

Some thoughts on using Zoom for services in a time of pandemic

First, some background: the COVID-19 pandemic took off during Lent 2020, and my parish had to make some decisions about how to do the services, and which services to do. Some services either need to be done in the church or it makes a lot more sense to do them there, while others can be done remotely, with nobody in the church. In general, if a service can be done as a Reader service, it can be done remotely, although there are some cases where a service involves something special like a procession that would be lost if it wasn't done in the church. To further complicate matters, two of our 4 chanters, including our Protopsalti (head chanter) are not allowed in the church during this time because they are old enough to be considered high-risk. We have also had a limit of no more than five "essential personnel" in the church; in our case those five were: the priest, an altar server, a chanter who also runs the live stream and sings alto, the choir director who also plays the organ, and one of the choir director's daughters who sings soprano. That leaves three chanters who have to participate remotely if they are going to participate at all.


We chose Zoom as the platform, in part because I had a paid account with no meeting length limit, thus no extra cost. Zoom is also capable of live-streaming. If you go to your Zoom settings and click "View more settings" it takes you to the full settings on their website. At the end of Meeting / In Meeting (Advanced) there is an option to enable live streaming, and a further option to enable custom streaming. I don't know if these options exist in the free version, but I have the lowest tier subscription and I have them. We use Facebook's "Create a live stream" page to create and schedule streams. If we're doing a service entirely on Zoom, I'll use the custom streaming option and give it the stream key; if we're in the church, I use OBS and give it the stream key. I always do a local recording in case there is a problem with the stream. It's usually pretty reliable, but every once in a while something goes wrong, and if it's serious enough, I'll post the local recording so that people can watch it later. If the service is being done entirely on Zoom, I tell it to do a local recording; if we're in the church, I do the local recording in OBS.


Doing a service on Zoom is a bit different. First, only one participant can be talking or singing at the same time, partly due to lag, and partly because the audio may break up as Zoom tries to figure out who is talking. However, if multiple people are on a single Zoom connection, those people can participate together. So, if you have multiple singers in the same household, they can function as a "choir", but singers in different physical locations cannot. Second, there is a lot of non-verbal communication that takes place during a service, and Zoom prevents that from happening. (Yes, there is a "chat", but trying to do that while you're in the middle of doing the service is difficult at best.) The way around this is to make a copy of the service text and assign parts, so that everyone knows what they're doing and when. This can be relaxed somewhat when some of the people are in the church, but any interaction between the folks in the church and the folks on Zoom needs to be scripted to avoid confusion.


Setting up the audio in Zoom also presents a bit of a challenge. At a minimum, you want to tell it not to automatically adjust the volume, and manually set the volume yourself so that it stays consistent. Also, if you're doing a service where some people are in the church and others are on Zoom, you'll need to have a speakerphone so that you can get the audio from the Zoom meeting into the church and vice versa without getting echo or feedback. I'm using Voicemeeter with auto-ducking on bus 3, which I've configured Zoom to use, so the microphone automatically mutes when there is audio coming from Zoom. The feed from that mic is also configured to go to to bus 1, without auto-ducking, which OBS is using for the recording and live stream. If you need a usb speakerphone, I recommend the Jabra Speak 410.


Although this may seem counterintuitive, sometimes it is better to have your Zoom participants connect their audio by phone. (Note that free accounts may not be able to allow this, but you're probably not going to want to use a free account anyway due to the meeting length restriction.) Make sure they have a calling plan that will allow them to do so without incurring toll charges before asking them to try this. If all they have is a cell phone with questionable reception, it won't work well, but I've found that when it works, I get a more consistent audio signal from the folks on the phone. Since we're not sharing video in the church to conserve bandwidth, the folks calling in don't bother going through their computers at all, except to download the service text. If you are sharing video, you still have the option of connecting your audio by either computer or phone, provided you didn't disable this when you scheduled the meeting. Calling in may also allow someone who is not particularly computer literate to participate - you'd be hard-pressed to find someone who doesn't know how to make a phone call and punch in a meeting number.


You may also want to make some adjustments to your parish sound system to make things easier to understand for both the live stream and the Zoom participants when you're doing services in the church. First, if the room has a lot of natural reverb, make sure you're not adding any more in the sound system. Too much reverb makes it hard to understand, and this is particularly true if you have to compromise on your volume settings so that all of your Zoom participants can be heard. Too much reverb may make the louder ones impossible to understand. You may also want to consider bringing your Zoom participants in on a speaker near the chanter's stand instead of feeding them into the sound system; if natural reverb is an issue, you may get better sound that way. Second, if you have any mics that sound at all boomy, turn the bass down a bit on those mics to make them easier to understand. More is not necessarily better in this case - too much bass can make it hard to understand, but if there isn't too much bass turning it down will make it sound tinny. Turn it down if there's too much, but don't go overboard. Third, if anyone, e.g. the priest, is wearing a headset or lavalier, and they're both going through the sound system and being picked up directly by the mic(s) you're using for the Zoom meeting and the live stream, they will become very hard to understand if they get too close to your mic(s). The distance at which this happens varies depending on the acoustics of the room and the mics being used; you'll need to test it, and if it's an issue, the person wearing the headset or lavalier will need to either stay farther away from your mic(s) or shut theirs off when they get too close.

How to do the Hours as a Reader Service (no priest)

You can find the Hours in the "Sacraments and Services" section of the AGES website (http://www.agesinitiatives.com/dcs/public/dcs/). [UPDATE: The Archdiocese has taken over the Digital Chant Stand; it's now at https://dcs.goarch.org/. The structure is still the same.]

Do the opening prayers (before "Come, let us worship...") if you are not doing the Hour immediately following another service. You can find these in the Ninth Hour. If you are doing the Hour immediately following another service, start with "Come, let us worship..."

Most priest parts are replaced by "Through the prayers of our holy fathers, Lord Jesus Christ, our God, have mercy on us and save us". You can find this phrase at the end of the Ninth Hour if you want to see it in Greek or want to refresh your memory. If a Priest part is preceded by some variation of "Father, bless", skip that phrase.

The second Priest part in the opening prayers, "Glory to You... Heavenly King, Comforter, Spirit of Truth..." is one of two that we read if there is no priest present.

The first part of the opening prayers is different during the Paschal season. You can see those changes in the first page of "Matins Ordinary", also found in the "Sacraments and Services" section.

The Apolytikia and Kontakia can be found in Matins in the daily services.

Sometime after the Kontakia, you will find "In the name of the Lord, Father, bless!" and a blessing from the Priest. These are replaced by "Through the prayers..." as described above. The Priest prayer following this is read.

The ending (starting with "Glory to the Father..." following the prayer described above) is not included in all of the Hours. You can find it in the Ninth Hour. It should be done if you are ending with this Hour; otherwise, skip it.

If you are doing the ending, skip "Father, bless!" and continue with "Through the prayers of our holy fathers..." at the end of the Priest's response.

Open Broadcaster Software (OBS)

A brief overview of my OBS configuration

Virtual Choir

General info for those recording parts

I'm not going to go through all the details of how to record your part here, however some people have been known to borrow their kid's computer and/or gaming headset to do their recording. A gaming headset seems to be a good choice - that's what I normally use, but note that every gaming headset I've seen that has both a usb and an audio connector uses the usb strictly to power the lights, and if you're doing a virtual choir recording, you probably don't want the light show. If you leave the usb connector unplugged, the headset will work normally, but the lights will stay off.

 

For those who are curious, this is how I currently record virtual choir parts. (If you don't basically know what you're doing, don't try to follow this - use whatever instructions you were given.)

  1. Import the guide track into Ardour. Ardour allows one video track, so this works for video guide tracks as well as audio-only guide tracks.

  2. Set up one track for each part being recorded, plus one extra track which I'll use to do the actual recording.

  3. If the guide track doesn't have obvious enough timing and it's feasible to do so, add a click track. (So far I've only had to do this once.)

  4. For each part, once I'm confident that I'm ready to follow the guide track precisely, record the part in the recording track.

  5. Line the recording up with the guide track. (I have Ardour set to nudge by one frame, so I can line it up precisely.)

  6. Slice the recording as needed to move the pieces that I'm happy with into the track for that part and re-record the pieces that I'm not happy with. (Make sure a "piece" isn't too small - recording just part of a phrase separately may sound weird. It's best to have a breath as a boundary at each end.)

  7. Export each part independently. I usually use either wav or mp3. (Some of the projects I've done have used Soundtrap, which seems to have issues with the sample rate in my wav files. It does the right thing with mp3 files, though.)

  8. If I need video, for each part, export a modified guide track that includes me singing that part in addition to the audio in the guide track, and the click track, if I added one. (Ardour can export a suitable video, if the guide track includes video.)

  9. Set up OBS Studio with feeds from the camera and the system audio. No microphone - I'm not recording my audio at this point. Also, no headphones - I can listen to the audio through my computer's speakers.

  10. Sing along with my previously-recorded part and record it in OBS Studio. This gives me a video with the guide track audio + my audio, and my video. Convert (remux) the video to mp4 before leaving OBS Studio.

  11. Open the video in OpenShot. Separate the audio and import the audio for my part.

  12. Line up my audio with the audio from the video, and mute the original audio once it's lined up.

  13. The audio may be slightly offset from the video due to different amounts of latency for different devices. If it doesn't look right, shift the audio as needed to line it up.

  14. Trim the video as needed and export it.

General info for those assembling the parts

You will need an audio editor and a video editor. I prefer Ardour for audio. I currently use a combination of OpenShot and Camtasia for video. OpenShot has more functionality in some areas, less in others. However, OpenShot allows precise track adjustments, while Camtasia requires you to drag things with your mouse and hope you get it just right. OpenShot also runs on Linux, so I'm currently leaning in that direction. I have seen recommendations for DaVinci Resolve, which is too buggy for my taste - it messed up my audio, left running processes behind on exit, and gave me a file with no audio when I added an effect that worked perfectly fine before rendering. I know many people happily use it, but at this point I am not one of them.

 

This is the basic process:

Audio:

  1. Import audio from all the tracks, organizing them by part.

  2. Line up all the tracks. Pick one track - the master is generally a good choice - and line up all the other tracks with that one so that your alignment point doesn't drift.

  3. Process each track as needed to clean it up.

  4. Balance the individuals within each part.

  5. Balance the parts.

Video:

  1. Check each video and make sure that the audio and video are lined up. Sometimes there may be some extra lag in the response of the microphone or the camera that will cause one of them to be slightly offset. If that happens, you'll need to separate the audio and video and shift it as needed so that when you use the audio to line everything up in the next step, it will in fact be lined up correctly.

  2. Import all videos, using the audio to line them up. Once they're lined up, you can remove the audio from all but one, which you will use to help you line up the final audio.

  3. Do whatever processing and/or placement is desired on the videos.

  4. Import the finished audio, using the audio retained in step 1 to line it up.

  5. Mute or remove the audio retained in step 1 and render the final video, trimming as needed.

Virtual Choir 101 seems to be a good source of information on how to do a virtual choir. Check out their guide "How to Create a Virtual Choir Performance".

Assembling the parts for a virtual trio

Zoom

How to connect phone audio in Zoom

This video is aimed ay my parish, but others may find it useful. Note that I'm using the desktop version of Zoom - if you're using the mobile version, the controls may be in a different place.

How to enable original sound in Zoom, and other audio settings for singing

How to stream a Zoom meeting to Facebook using a stream key

This video is aimed ay my parish, but others may find it useful.

How to use a phone as remote camera/mic for a Zoom meeting

This video is also aimed at my parish, but others may find it useful. The "other video" referred to is the one on "How to stream a Zoom meeting to Facebook using a stream key", above.