A quick guide to editing podcasts

This is kinda quick and dirty primer to making your podcast listenable. It will be focused on using Audacity, which is free and available on any platform. If you use something else, the advice should still be usable, but you’ll have to find the feature in your software.

First things first: Recording

A simple method for high quality recording when not sitting in the same room is to record in Audacity on each computer while using a voice chat software of your choice. The other voices send their audio (preferably lossless quality) after recording. This makes the voice chat quality a nonissue, so it makes my very bad internet service workable. Using time.is, each person can clap at the same time to make tracks easy to synchronize.

This one’s pretty self explanatory. Just hit record. When you are done hit stop. If your mic has adjustable volume, try to set it to have some headroom. The gauge with the green bars going wild shouldn’t reach all the way to zero. Loudish speech can probably be set to max around where I drew the line. You don’t want to see it hit zero, but staying close to there is good. Safer to go too quiet than too loud here.

Editing Time

So now we’ll get to editing. First is the sync, use the scissors icon (or ctrl-x) on highlighted sections to cut out recording before the clap. I leave the clap in case I think it needs adjusted while editing, then cut it out after I’ve finished editing. Once the tracks align pretty well, you should select Tracks → Sync-lock Tracks to keep things aligned while editing.

Now were going to match the volume of all voice tracks. It can be done by ear, using the slider at each track marked +/-. But I use the Loudness Normalization. Double-click a track to highlight the entire thing, go to Effects → Loudness Normalization. I find -18 dB works well for me in most cases. I do this to each track individually.

At this point if there is a bunch of background noise on the recording, it might be a good idea to make use of the Noise Reduction effect. I only use it on tracks with constant background noise. If you select like 10 seconds of silence on a track, then go to Effect → Noise Reduction and click Get Noise Profile. Now select the whole thing and it’ll try to filter out things that sound like the noise from those 10 seconds. It works shockingly well in my experience.

Next is a compressor. In my opinion, using a lot of compressor on this sorta thing is best. It makes it easy to listen to at lower volumes or in louder environments, like in a car. Here’s what my typical compressor settings look like. You don’t want the compressor to capture noise, so the noise floor can be tweaked to make sure that happens. All of the speech you want to fall above the threshold setting (at it’s peak). Use the Solo button to listen to only 1 track and watch that green gauge again to get an idea where these need to be set.

Next I use a noise gate. I set it to cut quiet sounds down to silent. Getting too aggressive with this sounds like crap. This makes quiet parts silent, so when you make cuts they’ll sound more seamless. Listen to some quiet part of each track and decide on a cutoff. If you can hear the voice call in the recording, it’s a good idea to try to get this to cut it out, so you won’t hear it with the delay in the final cut.

There is a bug in the noise gate plugin that causes it to not work if you select too much audio. It seems to work on up to 40 minutes at a time, so if you have more than that selected it will seem to be working on it, but won’t change the track at all. I do each track in about 40 minute chunks, checking to make sure it applied along the way. This is the most annoying part, but ctrl+R will reapply the last used effect.

Consider using an equalizer, maybe download a de-esser plugin if your s sounds are harsh on a person’s mic. I haven’t been doing this with my current setup, but you may need to.

Here’s an example of my settings for that one:

Now we’re going to automate away a lot of small cuts where there’s unwanted silence. Use ctrl+A to select all tracks then select Effect → Truncate Silence. I find people don’t generally pause for effect for much longer than 3 seconds, so it won’t generally hurt. I usually only lose a few minutes off a 2 hour recording. If you lose a big chunk something probably went wrong.

And finally I get to the part where I just listen to the whole thing through. I think this is important, but if you are just all rock-solid speakers, you may be able to get away with skipping it. I’m always doing some light editing to keep the flow of conversation going a little smoother. You have 2 main tools for this one, the scissors (ctrl+X) and the silence selection tool (ctrl+L). Cutting will get rid of time, silencing will help if people are talking over one another, or if one person is typing or taking a drink. Using these two together you can do most anything.

Special Effects

I like to do intro music and outro music. You may want a segment transition too. Copyright considerations are outside the scope of this write-up. There are many royalty free tunes on the ‘net, and talented people who will do a commission for theme music if that tickles your fancy.

You’ll likely have to turn off sync lock tracks. When moving tracks around, if you see a yellow line appear, it’s aligning the cursor with the start or end to each track. So I put in my audio, then move it to the beginning, move my vocal tracks forward, then cut the clap that I left in at the beginning. Check out the adjustable fade effect for this kinda stuff.

Now I export as a simple high bitrate MP3, because I don’t think there’s much point in delivering a podcast in higher quality than that. I upload that to anchor.fm, but there are other options to blast your podcast out onto the internet. Anchor works quite well for me though.

Some Quick Tips:

Guide by: @timmy@jorts.horse

Feel free to reproduce in whole or in part as you wish. Attribution would be polite but is not required.