cssquirrel/how-we-record-stardate-supplemental.md

## how-we-record-stardate-supplemental.md

      
    Raw
  

              how-we-record-stardate-supplemental.md
            
          
    How We Record Stardate: Supplemental

Equipment and Software

Each of the Brothers Weems records using a Blue Snowball microphone and Audacity audio software. We also have the free LAME encoder which allows Audacity to export to MP3.
Recording Procedure

We're in our own apartments as we record, each recording our own stream while on Skype together. To later make it easier to sync our tracks, one of us plays a loud buzzer while the other holds headphones to their microphone. Once we're done recording, we export our own soundtracks to MP3 format with 128 kbps, stereo, constant rate sounds.
Post-Recording Editing Process.

Step 1: "Condition" Each Brother's Track Independently.

Due to volume and background differences between the brothers' recording locations, the "conditioning" process should be performed separately on each track to minimize unwanted sound quality issues from making it into the final product.
For each track, go through the following steps in order to match the overall process Kyle has used when sound editing.

Import a brother's track. File > Import > Audio. Choose the raw file from the brother you are editing.
Select the entire track. Choose Effect > Limiter, then Hard Limiter type and Limit to (dB) -0.50. (I don't touch the other settings). This fixes "clipping", which is where the sound on the mic clipped against it's maximum threshold. (Kyle in particular is pretty loud and does this a lot.)
Select the entire track. Choose Effect > High Pass Filter. Set the Frequenxy (Hz) to 80.0 and the Rolloff to 48 dB. This will remove all sound below that threshold, which is below the human range of speech and helps cut many unwanted low sounds.
Select roughly five seconds of "silence" in the track when the brother isn't speaking. Ideally a relativey "noisy" bit of silence where background sounds are on the louder end of average. Choose Effect > Noise Reduction, click Get Noise Profile. Then select the entire track and use Effect > Noise Reduction, click OK. This will perform noise reduction that helps remove a good deal of background noise from behind the track to account for the "silence" of the room.
If you are editing Conrad's track, go to the beginning of the track and zoom in to find the synchronization buzzer effect. It's particularly loud on his track as the source of it comes from his end. Select the buzzing sound. Choose Effect > Clip Fix. Set Threshold of Clipping to 95. Set Reduce amplitude to allow for restored peaks to -4.00. Do this twice. By doing this we'll make sure the buzz sound is roughly middle of the track in loudness so it won't distort the compression and normalize steps.
Select the entire track. Choose Effect > Compressor, I set the settings as Threshold: -18, Noise Floor: -40, Ratio: 2:1, Attack Time: 0.5 seconds, Decay Time: 1.0 seconds, and leave both checkboxes unchecked. I don't know enough about any of these settings to comment much on it, other than to say that works for me. Using the compressor helps bring the loud and quiet sections closer to each other, which helps overall perceived quality of the speaker's voice and lifts up the quieter moments of talking to not be too low for listeners.
Select the entire track. Choose Effect > Normalize, selecting Remove DC Offset, Normalize Maximum Amplitude to -1.0 dB, and I do not check normalize stereo independently. This "normalizes" the volume of the track so that it's highest point is at the indicated decibels. Doing this will make sure each brother's track is roughly the same perceived loudness, instead of one being much louder than the other.
After these steps, there might be new incidents of clipping, so repeat step #1.
Export the track to a new "Processed" track: File > Export > Export as MP3. Name the file "NNN-Name-Processed.MP3" with episode number and brother name (e.g. "042-Kyle-Processed.MP3"). Use leading zeroes for episode counts below 100. Set Bit Rate Mode to Constant, Quality to 128 kbps, and Channel Mode to Stereo.

Step 2. Combine Conditioned Files Into A Single Track.

In a new Audacity window, do the following:

Import Kyle's track. File > Import > Audio. Select Kyle's conditioned MP3.
Import Conrad's track to do the same.
Go to the start of the tracks and zoom in until you can see the buzzer sounds. It will look bigger on Conrad's track than Kyle's. Cut off the start of Kyle's track to roughly a second before the buzzer sound begins. Then click on the track at the very beginning of the buzzer to create a line.
Select the "Time Shift Tool" (it is an arrow pointing left and right in a cluster of buttons to the right of the record button.) Grab Conrad's track and line up the start of his buzz event with Kyle's. If you do this at a high zoom, the tracks will line up very well with no perceptive desync.
Export the tracks into an unified combined MP3, using the same instructions as point #9 of the process above. Name it "NNN-Combined-Processed.MP3" with the appropriate episode number.

Step 3. Add bumps, stingers, and background music.

Now that we have a conditioned, combined track with both brothers, we're ready to start adding bumps, stingers, and music as well as editing out any content that's not intended for playback such as bathroom breaks.
To setup up for this process, in a new Audacity window, do the following:

Import the Combined track. File > Import > Audio. Select the Combined MP3.
Add a label track for keeping notes if needed. Tracks > Add New > Label Track.
Add a stereo track that will hold background music and sounds (stingers and bumps that interrupt the brothers, etc, go in the primary track). Tracks > Add New > Label Track.

Time to start editing! This is now an artistic process, not just paint-by-numbers. What follows are general guidelines, with some specific guidance for specific stinger/bump inserts or standards for ongoing bits.
The Opening

Conrad prefers cold opens, similar to the Comic Place's Perfectly Acceptable Podcast, where the podcast starts partway into a conversation at a humorous spot. Kyle prefers the opposite, with an explict "Welcome to Stardate Supplemental..." start.
As a result, sometimes we do one, sometimes we do the other. We'll decide between ourselves which we're doing. If it's a cold open, you'll need to choose for yourself a good spot to open in the first few sentences from us warming up after we're clearly doing pod content. Let the brothers ramble for up to five minutes (no later) before inserting "ss-bump-intro.mp3".
If it's not a cold open, there will be a specific indication like "Kevin, we're starting here" followed by Kyle giving the opening spiel. In this case, insert "ss-bump-intro.mp3" at the very start of the track.
From the start of the opening segment when the brothers are speaking, insert the background effects of "ss-background-for-segment-1-(enterprise-d-bridge).mp3". This lasts for 3 minutes, but you can fade it out early if needed (such as interrupting for a Stardate Supplemental opening bump during a cold open.)
Segment Transition Bumps and Backgrounds

At some point we'll transfer into the second segment where we discuss the ship. We insert "ss-bump-for-segment-2-(intercept-course).mp3" in between, and then play "ss-background-for-segment-2-(voyager-bridge).mp3" in the background, which lasts 1:30.
We then later transition into the third segment "What Would DeSoto Do". We insert "ss-bump-for-segment-3-(wwdd-desoto-on-comms).mp3" in between, and then play "ss-background-for-segment-3-(tng-command-center).mp3" in the background, which also lasts 1:30.
After we finish our What Would DeSoto Do bit and say "And credits...", we insert "ss-end-credits-by-conrad.mp3".
After the closing remarks of the final segment, insert "ss-holodeck-outro.mp3" on the very end of the track.
If the brothers ever go away mid-segment to deal with a cat or the like, cut out the interrupting material and insert the "ss-elevator-music.mp3" bump. As a gag, Kyle speeds that bump up by 20% every subsequent time he has to use it in a given episode, but that's not required.
Background Music For Specific Bits

Vulcan Mystic Stuff: When the brothers are discussing the Vulcan soap opera "The Trials of the Sinister Kuvok" or Vulcan mysticism/weird pajamas, play "ss-bit-background-(trials-of-the-sinister-kuvok).mp3" in the background, fading it in and out as required to fit the bit length.
Episode Recap: If the brothers start reviewing the plot of an episode (instead of the details of the ship of it), use the accordian background music, starting with "ss-summary-accordion-loop-fade-in.mp3", ending with "ss-summary-accordion-loop-fade-out.mp3", and filling with repeated filler of "ss-summary-accordion-loop-loop-10sec.mp3" or "ss-summary-accordion-loop-30sec.mp3" depending on how much space you need to fill behind it (sometimes we waffle on for quite a while).
Closing Details: When Kyle is going through the closing spiel, starting with "You've been listening to Stardate Supplemental...", play "ss-bit-background-(closing-remarks).mp3" behind them, fading in and out to fit length as needed. 30 seconds before the final words where Conrad says "Byeeeee", replace the background music with "ss-bridge-background-30-seconds.mp3". Fade out the closing remarks music prior to the insertion of the bridge background music.

In addition to effects and bumps, I do a good deal of cleaning of the track by editing out many "ummms", long pauses, or other odd sounds. Conrad's leaky-nosed Oz can be heard making strange snort noises at times, but often he does that behind Conrad's talking so those slip through.