Among the variety of nonsensical things I do on the side is record a podcast. As a host, I’m not that great - we’re into the teens now as far as an episodes setlist is concerned, and we normally sit in the top 20 for our category of podcast on iTunes. And, while I think I’m getting the hang of the technical side of it all, I’m still working on the performance side - I can do multi-day workshops, I can train, and I can do the whole thought-leadership thing, but being entertaining for 20 minutes is hard. I have far more respect for comedians now.
But, that’s not what this post is about. Instead, it’s about some basic pointers to improve your recording quality (with pictures and explanations!). To be fair, this is seriously basic stuff - it assume negligible technical equipment, and is more focused on getting the source right than anything else.
Recording badly is irritating at best, a downright turnoff at worst. While I don’t have any numbers to support this, it’s my belief that the quality of recording in a podcast is directly proportional to how professionally it’s perceived. Sure, you may talk crap, but making it sound good is probably the easiest way of building credibility. If you ain’t got it, fake it.
The final quality of a podcast is in general, directly related to the quality of the source material. You can make a crap recording sound OK, but it’s a hell of a lot of work, and it’ll never sound as good as something that was well-recorded in the first place. In general, getting a good source is pretty easy. In theory, anyway - you just want to:
- reduce all background noises as much as possible
- maximise the volume captured while avoiding clipping
- maintain a consistent volume level
That’s it! You can spend a lot of money on expensive equipment, but if you can’t get those three things right, everything else is a waste of time.
Eliminating background noise in your standard computer room is fairly easy - just make sure you do the following:
- shut the door
- keep the microphone (and microphone cable) away from any wireless equipment or other sources of interference
- try and keep the mic as far away as possible from any fan noises (such as from the computer)
Maximising the volume captured while avoiding clipping is a little more complicated. First, a bit of background. Every microphone / input has a maximum range of volume it’s able to capture. Below a certain point, the microphone / input isn’t sensitive enough to pick up the noise. Above a certain point, the noise is too large for the microphone / input to handle. With the first, you hear nothing. With the sound, you end up with massive distortion.
How does clipping happen? Generally, you’re either talking too loud or, more commonly, you breathe on the microphone. That wind translates to massive energy, which creates a great deal of clipping.
Why is this important? Well, clipping generally makes things sound like crap (overdriven guitars being one of the notable exceptions) - you hear lots of pops and distortion. However, there’s other reasons why clipping is bad when you’re doing a podcast. Many recording programs (Skype included) will automatically adjust the recording volume for you. So, when you breathe on the microphone, it thinks you’re being really loud, so it drops the volume on the input to accommodate. Unfortunately, because you aren’t, the volume’s now too low, so everything you say for the next few seconds are almost inaudible. A double-whammy, if you will - you’ve make it sound like crap, and to accommodate for the noise, your computer effectively cuts you off.
Fixing all this is actually pretty easy, it just takes some discipline. First of all, don’t breathe on the microphone. Ever. The easiest way to do this is to put the microphone below your chin (making sure it’s pointing at you), roughly one to two hand’s distance away from you. When you talk, just talk over the microphone, not at it.
Next up, don’t move. Hold your position, even if it kills you - every time you move, you modify the volume of the input. A little motion is probably inevitable, but whatever you do, don’t lean back and forward - you’ll completely screw up your recording.
If you want to get the *best* source possible, try recording a test piece and have a look at the waveform. Like these, that I’ve prepared earlier!
Here’s a good waveform:
You can tell it’s good because of the following:
-
the waves are large and have some variation
-
they don’t hit the top and bottom of the bar (a representation of clipping)
-
the background line is fairly straight, suggesting relatively little background noise (in my case, my fan’s very irritating)
Now, here’s a bad waveform:
It’s bad for the following reasons:
- the speech parts of the wave are really small - they’re quiet compared to everything else, forcing a great deal of additional processing
- there’s a massive amount of clipping on the leftish side of the picture - in this case, it’s because someone breathed on the microphone
If you want to tune your setup (which you most definitely should do), just try recording a test piece. If it looks like the good waveform, you’re in good shape. Try talking at the loudest and quietest volumes you think you’re likely to be talking at - if there’s clipping, move your microphone away from you or drop the recording volume. If it’s all too quiet, move it closer or bump up the recording volume.
Then, when you’re recording proper, just remember these things:
-
don’t move
-
don’t breathe on the microphone
Easy!
Dynamics!