It was Tuesday, November 24, 2020, and I was piecing together audio from the previous weekend’s podcast recording session. As always, I opened up my own Audacity file and imported my two co-hosts’ audio files. Then I downloaded the mp3 sent to us by our guest speaker, planning to add his voice into the mix, just as I had dozens of times before.
Immediately upon importing our guest’s audio file, I could tell we had big trouble on our hands.
My first clue that something was off was that our guest’s audio file was not the same length as the other podcast participants. Since we always sync our start and stop times, that was concerning. I could also see a ton of static and noise throughout his waveforms.
But then I pressed play on his audio and my stomach dropped.
It was bad. Maybe even unusable.
He sounded like a malfunctioning robot who was also fully submerged underwater. Sure, his audio had been somewhat bad during the recording session, but not this bad.
And the worst part was that, somehow, it seemed as if the whole file had been sped up ever-so-slightly, meaning that none of our conversation was syncing up the way it had actually occurred.
I looked at the clock. Roughly six hours until the episode was due to release.
It was at this point I realized that I had some tough decisions to make, and a LOT of work ahead of me.
Uh oh.
How Could This Happen?
Let’s face it: bad audio is not rare.
In fact, it’s probably more common than good audio. But as podcast editors and producers, our job is to make a listenable show. That includes having some degree of control over audio quality, even if the recording of said audio is not happening in our own physical space.
If you found this article by frantically Googling how to fix bad audio, and fast (like me on Nov. 24, 2020), you’re probably not all that interested in hearing how to prevent bad audio from being created in the first place.
Let’s not dwell on the past, after all. There are plenty of resources you can read up on before your next recording session. Instead, let’s focus on the bad audio we have on our hands and, more importantly, what to do about it.
Dealing With Bad Podcast Audio: What to do?
If you fill the roles of both editor and producer of your podcast, as many DIY podcasters do, this is the point at which you need to make an executive decision.
Is the audio salvageable, or should you scrap this episode?
The answer will come down to two things: how fixable the audio file in question is, and how much time you have (or, more realistically, how much time you’re willing to invest into fixing it).
Trying Again
In a perfect world, you could simply reach out to your guest and co-hosts, explain the situation, get everybody back together, solve the audio issues, and have another riveting conversation, this one even more spirited than the last.
But let’s face it, there’s probably a million reasons you can’t do this.
Coordinating schedules is never easy, and even if your guest is willing to come back and do it again, you may not feel comfortable asking them to give up even more of their time to essentially have your previous conversation a second time. Plus, your second attempt will never have quite the same magic as the first.
Using Second-Hand Audio
If you record in person with your podcast partner, it’s possible that some of their conversation may have been picked up by a room microphone, another host’s mic, or even yours. If you record remotely, you may also capture a recording of your VoIP phone service directly in addition to editing together each participant’s local audio.
In cases like this where you happen to have additional recordings of your conversation, it may be possible to grab the audio of the person whose equipment malfunctioned from another person’s mic. With a careful volume rebalance and some clever editing, you may be able to recreate a somewhat clear audio track from one of these ambient sources to replace the distorted audio file. This will not be fast or easy, but if you’re determined it could be an option worth considering.
In some cases, I’ve heard of podcast hosts redubbing their own audio upon learning that their microphone had failed to capture their local recording. If your podcast host is a talented voice actor you may be able to get away with this. But for the rest of us, this is probably just as unrealistic as organizing a second recording session.
Fixing Bad Audio
Before you can make the decision to scrap or save this episode, you need to have an idea of how good you can manage to get the bad audio to sound. There are a number of things you can try to fix bad spoken word audio.
Addressing Sound Quality
- Noise Reduction: Many audio editing programs have a function that will allow you to isolate a bit of silence (aka “room noise”) from the track in question, then highlight the full track and attempt to remove that sound from the entire track. This is a good place to start when trying to salvage audio.
- Remove Pops and Hisses: Similar to the Noise Reduction effect, some programs like Adobe Premiere have effects that specifically locate and eliminate hisses and other audio distortions.
- Mute spaces between speech: This could be time consuming, but if you’re committed to salvaging audio with excessive background noise, this may be what’s required, especially if track alignment has been distorted (more on that below).
- Adjust the equalizer: If the faulty track is acting up in the high or low end of the mix, you can run an equalizer on the track to try to lessen the impact of those frequencies. For instance, if the track is most staticy when your guest’s voice goes low, consider turning down the bass.
- Removing noise with the spectrogram: This starts to get pretty advanced, but if you have a consistent whining pitch throughout your audio, using the spectrogram to remove that frequency range could help to eliminate it. This video does a good job walking you through the process in Audacity.
Addressing Alignment
- Stretching or Shrinking the Track: If the conversation is unaligned, it’s possible the file somehow compressed or was stretched during export. A clue to this can be if the person’s speaking voice sounds slightly higher (chipmunk-ified) or lower (slow-mo voice). If you synced up your start and stop times during recording, you may be able to select the whole track and compress or stretch the entire thing to align with the uncorrupted tracks. In my case this didn’t work due to other distortions, but it’s worth a try.
- Realigning speech: If your guest’s audio track still isn’t lining up like it’s supposed to, you may need to mute spaces between their speech as outlined above and then shift the beginnings of sentences to the correct positions. This will require listening as you edit to make sure the conversation makes sense. Be extra careful not to throw off the alignment of the good audio tracks.
Making the Judgement Call
At the end of the day, whether or not you decide to release this episode is completely up to you. But here are some factors to consider as you mull over your decision.
The Conversation is the Most Important Factor
You wouldn’t even consider publishing an episode with bad audio unless you really felt like the conversation that took place was worth listening to, right? Chances are your listeners would agree. The vast majority of podcast listeners will soldier through a sub-par audio experience of a show they like as long as they’re convinced it won’t be a regular occurrence.
Check In With Your Guest
If you’re not quite sure what to do, you may want to check with the person who recorded the bad audio, especially if they were an honored guest on your show. It’s entirely possible that the thought of appearing on your show with distorted audio is not ideal for them, and they may want you to scrap the episode.
They may also be just as likely to be fine with it, especially if you’re able to restore it to an acceptable quality level. Either way, be sure to make the party in question a part of your decision making process.
Prepare the Listener
Either at the beginning of your show or the beginning of the corrupted segment, edit a quick editor’s note into the episode. There’s a few things you might want to mention during this disclaimer.
- Apologize for the bad audio coming up
- Let them know how long it lasts in case they prefer to skip past it
- Assure them that it’s still listenable
- Confirm that this is a one-time thing and that you’ll prevent this from happening again in the future
I would also recommend adding a sentence or two to your episode description. This could actually get even more people to listen than usual, as they may want to skim through the episode out of morbid curiosity. “Just how bad is this audio anyway?” they’ll ask. “Let’s listen and find out!”
Get Creative
In the case of my underwater robot guest, one of my co-hosts helped to salvage the situation somewhat by using a voice changer on his own voice during the intro of our show.
Then, when I edited in my disclaimer, I jokingly mentioned that we had accidentally applied a voice changer to our guest.
Did this help the audio quality get better? Not even a little bit. But I like to think that it did succeed in setting the listener’s mind at ease and opening them up to the possibility of listening to the segment in spite of the subpar audio quality.
This Too Shall Pass
In the end, we made the decision to publish our interview with our guest. To date, it’s one of our most downloaded episodes, and the guest in question has adopted a robot emoji following his screen name.
No matter what you decide to do with the bad audio you currently have on your hands, just remember that your next chance to impress your listener base will come with your next episode. Bounce back, learn from your mistakes, and prepare to do it all over again (but better) during the next recording. You listeners will be there, so make sure you show up too.
About the author
Matt Nadolny is a digital marketing specialist during the week and a podcaster on the weekends, known to his listeners as Tuesday Timp. He is the founder, producer, editor, and one of three co-hosts for the bi-weekly video game community podcast Go Mode: A Link to the Past Randomizer Podcast. Matt resides in Charlotte, NC with his wife, newborn daughter, and two cats.
