Workflow for audio processing for podcasts

Special Interest Groups Podcasting
1

Hi All:
I’ve gone through the mix music and voice for podcasting tutorial to record a short standard intro for my podcast. Well-presented, but I have a few questions concerning the workflow on using compression, gain adjustmnent, and normalisation. The workflow in the tutorial is:

  1. Record narration
    2a. Edit narration.
    2b. Use compressor on narration

  2. Import music

  3. Time shift tracks

  4. Use Envelop or Autoduck to reduce the music under the narration

  5. Fade music before and after narration

  6. Adjust gain on both tracks if clipping occurs

  7. Save and Export. Optional: Mix and render and Normalise

  8. In my case the imported music track has some clipping. The tutorial compressed the voice track, but not the imported music track. Why is that?

  9. The tutorial has Mix and Render and normalisation as optional. Why would you not do that?

  10. I will pre-pend the intro to each narrated podcast. The podcast will be a straight audio recording. I’m unclear on how compression and normalisation should be used in that case. I can think of:
    (a) Compress the podcast track only. Prepend the previously compressed and normalised intro. Normalise the rendered mix again.
    (b) Compress the podcast track only. Prepend an un-normalised intro. Normalise the rendered mix again.

In short, I’m not sure how much compression, gain adjustment and normalisation (and other processing) should be applied to separate tracks before they are mixed together; and how much to the final mix.

Thank you for your time.
Garry

2

What’s it sound like? If you’re getting commercially available music, it’s already been compressed and squashed and densified to death. Your live voice will tend to be light and airy, expressive and real. in short, it won’t match the music at all. Processing is likely to be needed on the voice rather than the music, but it all depends on what it sounds like.

You should be listening to the mix on a good sound system (or headphones) and judging the work on that rather than automatically applying a laundry list of effects and filters. If your work is intended to come out of a cellphone, you should make a pass where you listen on one of those, or something that sounds like one of those.

This is where you give us the address of the podcast so we can hear it. We can’t judge the work from the laundry list any more than you can.

Koz