MediaSlap blog

In part one of this article I discussed the definition of podcasts and their promotion. Since we published that I’ve had requests for some information on how to create an audio podcast. This is more of a simple overview of the mechanics involved in creating a podcast rather than how to structure content. I suggest you have a script, outline, or interview questions prepared before you record.

Use good microphone technique
Record each person speaking on the podcast with good quality directional microphone that will complement the persons voice while not capturing background noise. Your recording should be done in a quiet place that is not reflective. Unless the subject is spelunking you don’t want a room that sounds like a cave. Use windscreens on the microphones to reduce breath noise and pops if necessary.

Use a digital recorder
Digital is quieter than analogue and it will save you a bunch of time because the end product needs to be digital. You can record on your computer or use a CD recorder. If something goes wrong and you won’t be able to record the program again use redundant recording methods. For example you may record the program on both laptop and a CD recorder simultaneously. When using a computer to record, only run the recording program – resist the temptation to play games at the same time.

Edit your program
Clean up the resulting recording by cutting out extraneous words and stutters. If you have time constraints, edit out unnecessary words to reach the desired program length minus the length of any prerecorded open and close. Be careful when editing. Don’t edit in the middle of a breath or leave vocal artifacts – you don’t want it to sound edited. Then add your open and close to the program. Remember you need to have rights to any music you use to avoid copyright problems. Use a program like Sony Sound Forge or Apple’s Garage Band to edit and mix.

Mixing and processing
After editing you must mix or render the final product to a stereo or mono file. I usually process the recording and mix at the same time. First if necessary, use software to reduce noise. Then normalize the sound file levels (this is not compression) and add any EQ to make the tonal qualities more desirable. Finally compress the overall program and peak limit the file to -.3db to prevent digital distortion. The EQ, compression and peak limiting processes are generally over used and will ruin the audio. If we were talking food here, a little salt can bring out the flavor – too much and it’s disgusting.

Our end product will be streaming format(s) like MP3, WMA, Flash or MOV. The smaller the file the faster the listener can experience the podcast. But smaller files mean lower quality. Choosing the right rate depends on your audience and their connection speed. You can deliver a better sounding podcast to a DSL or cable audience as opposed to dial-up, it’s a compromise. For audio I usually use 22,050 Hz, 16 Bit, Stereo. One minute of audio compresses to a file size of approximately 0.45 MB (about 22:1 compression).

Choose a good podcast host
Don’t pay too much for bandwidth. More listeners need more bandwidth and a live event will require more bandwidth than on-demand. Multiply your estimated number of listeners by the estimated file size to get an idea of your needs. If you are just starting out I recommend using a host like www.libsyn.com, www.christwebs.com. For huge audiences use a service like www.akami.com. Akamai uses several servers in different geographical locations – listeners are routed based on both location and server load.

Make a good impression
Do it right or hire a professional to help you. For some ideas on how to promote a podcast see our article “Everything you ever wanted to know about…Podcasting! (part one).”

###