The State of Podcasts & Vlogs

The social and cultural message is the driving force on the web today. Ok, fine. But, why ruin your show with fundamental problems? I'm not talking about attention to detail... I'm not even talking about beginner errors!

What Bites Today? Rabbit Bites. Most of their shows are riddled with audio distortion and volume levels that dramatically change from scene to scene -- so much that it ruins the entire production. The viewer gravitates to the errors not the message. Heck, sometimes the volume is so low you can't hear those precious bunnies put down their prey in their cute high-pitched voices. Ever hear of a "pop filter"?

Some advice:

• Keep volume levels consistent. Reference levels for the web do not exist, as far as I know (please tell me if the W3C has 'em!) I suggest using -18dBFS[1] reference for your web content. After recording, you might need to run a compressor on your signal to smooth it out. Most modern audio editors have them built in. Don't know how to use it? Comment below and let's figure it out for you.
• Use a pop filter to reduce plosives[2]. They are inexpensive.
  
• Use the recording meter(s) to maintain a level that isn't too low but avoids clipping[3]. Check the complete recording chain to make sure gain is maximized at each stage. If you are using consumer equipment, switch in the AGC (automatic gain control.) I hate them, but at least you'll avoid clipping.
• Try and keep the acoustical properties matched within a scene. Don't change mic positions unless there is a good reason within the same scene (for instance, at a camera change.) The result is that the background ambience (the acoustical fingerprint of the scene) changes from scene to scene and becomes distracting.

Do you have any examples (good or bad) of podcasts or vlogs? I'd like to create a list of examples to help others in the pursuit of the technically perfect cast. Please post them below.
----

[1} "dB" or "decibel" is a relative term expressing the ratio of two levels. When speaking decibels, you must reference what you are measuring against. In our case above, "FS" means "Full Scale" or in practice, a signal at the highest digital level. Thus -18dBFS is 18 decibels down from full scale.

[2] plosives are produced by sudden airflows across the diaphram of the microphone and muddy the sound.

[3] There are several reasons for clipping. Most often, the volume is set too high somewhere along the recording chain. This results in clipping off the outer boundaries of the audio signal and results in distortion. Here's a closeup of clipping:

Audio Clipped Close-up

For more about personal multimedia, see my site: brainFlakes.