Fixing A Quiet Voice In A Skype Recorded Podcast Episode

I was talking with someone on Twitter recently, about an issue they had with a podcast episode. It turns out the Skype call recorder they used had set up the audio rather strangely.


Now this was odd to me as I’ve always had my Skype recordings split left and right appropriately. So I dug in to check it out. I grabbed a copy of his episode, opened it in Audacity and saw this:

NewImage

How very strange, indeed! The audio wave form shows me exactly what Jay said he was hearing – his voice plus his guest on the left, and the guest voice only on the right! Honestly, this baffled me. I’m not sure how this happened. I’ve never had that happen when recording a Skype conversation.

But, no matter! The goal wasn’t to fix the Skype recording setup and re-record. The goal was to fix this one episode, as is.

Fixing The Episode

After listening to the episode, I found a few things that were important:

  • Jay’s voice was significantly quieter than his co-host
  • Jay and his co-host did not speak a the same time

That second point was actually a saving grace! Because they did not speak over each other, there was a simple solution to this problem: 

  1. Split the stereo track in to two mono tracks
  2. Drop the track that was only the co-host
  3. Compress the remaining track to level out the volume of both people

Step 1: Split The Audio Tracks

This step is one that I have talked about before: splitting a Skype call recording into multiple tracks, and exporting as mono.

You should read that blog post for the complete steps on doing this. 

Step 2: Mute The Single Speaker Track

Audacity makes this easy. You just need to click the mute button on the track in question. 

NewImage

Alternatively, you can completely delete that track using the “X” button in the top left of the track. Either way, the result is that this track no longer plays – exactly what you want. 

Step 3: Compress The Remaining Track

With Jay’s voice being quieter than his co-host, I want to compress the audio so that the loud parts are quieter and the quiet parts are louder. Note that this is not to be confused with file compression or MP3 compression. This is audio compression, as I discussed in my article on file and audio compression, and why you need both.

In Audacity, I opened the Compressor from the Effects menu

NewImage

I pretty much left the settings that were there already, except for the “Make-up gain for 0 dB” checkbox. I unchecked that one, because it can introduce unwanted clipping and other issues. 

NewImage

Hit OK, and let it do it’s thing! It will make Jay’s voice sound louder, and his co-host will be slightly reduced at the same time.

The Final Episode Sounds Better

The net result is that the podcast will sound more even between the two speakers. This can be seen in the audio waveform, and heard when playing back the track.

NewImage

Now you can export this as a .mp3 file again, and the episode will be significantly better sounding, over all. You won’t have the strange sounding problem of one voice only coming from one ear, and both coming from the other. 

Hopefully the original problem with Skype call recording can be fixed. But if not, at least there is a way to fix it!

 

 

P.S. Check Out Jay’s Podcast

In the process of listening to this, I found that I like the direction that Jay and his co-host are heading with their podcast. His blog post in general seems pretty solid, and I like what he is doing with the JVMChat and community. You’ll definitely want to check out his podcast and blog