“Let’s do a podcast” I said to Chris back in around September 2016. Slightly surprisingly he said “Yes”. The adventure began.
This is a post I’ve been meaning to write for a while, but only now spurred into it by a chance tweet. It describes technically how we put WB-40 together every week. The editorial side of things is a separate topic…
First of, the setting. Chris lives in the West Midlands, I’m in South West London. We have only once recorded the show in the same physical location. The past year and a half has been an exercise is how we can record the show with as good sound quality as we can, but without spending a lot of money.
It’s also worth at this point noting my prior knowledge. For years I’ve dabbled with electronic music in one form or another. In my university days I was also pretty involved with student broadcasting (being one of the people who founded the Student Radio Association no less). To the extent that I learned my radio craft it was back in the days of using tape and razor blades to edit. I’m also fairly certain that I’ve learned quite a bit from my dad, who was for a period of his career a broadcast engineer at the BBC.
We’ve experimented over the time with a few different audio streaming services, but keep coming back to Google Hangouts. It’s not infallible, the sound quality is a bit ropey, but it’s fairly stable (or at least as stable as BT and Virgin Media’s archaic networks allow). However, it’s not just a matter of recording straight outta Hangouts. The recording setup is slightly complex…
At my end of the setup there is a reasonable quality Marantz mic (about £50), an old Yamaha mixing desk (about £100) and a USB sound card (Behringer – £25) that gives a line input into my laptop. I then use an iPad for connecting to Hangouts, with a line cable outputting audio into the mixing desk. When we record it means that I’m recording at good quality direct from my mic, reasonable quality out of Hangouts, and Chris is hearing me through the mic on the iPad. The final bit of the equation is then that the Mic and iPad outputs are panned hard left and hard right. Which in turn means that I’m using Audacity to record in “stereo” but effectively as two independent tracks which means in the edit I can take out any background noises (aeroplanes over Teddington, Chris’s exceptionally squeaky chair in Tamworth, occasional echo that seems to seep into Hangouts).
As we talk I record using Audacity (wonderful freeware that I’ve been using since the previous tool of choice Cool Edit Pro was acquired by Adobe and made ludicrously expensive as Audition). Once we’ve finished I then edit.
There’s a bit of music that was self composed and recorded using a package called Caustic on an Android phone. Amazing what you can do with a phone these days. It’s not going to win a Novello award, mind. The Theme tune is about 60 seconds long, and the spoken intro (and outro) goes over that. There’s an Audacity feature called AutoDuck which processes the music track to dip down in volume as the voice track requires.
For each segment of speech, the following is needed: first off, trim the front of the track to remove any silence or mumbling. Then align the pair of tracks to the right starting point. Then convert the stereo track to mono. For my audio track I then use compression (an audio technique to smooth out some of the variation in volume) in Audacity. This is mostly because the Hangouts audio is automatically compressed quite heavily, so applying similar to my voice makes it sound a bit more balanced. I adjust the amplification of Chris’s track to get to a similar point to mine. I then silence any quiet bits (selecting them manually by eye – there is an automatic function to do this that I must learn), and then pan the tracks slightly to left and right so there’s some stereo separation in the final mix.
The two mono tracks are then “Mixed and rendered” to use the Audacity lingo (basically turned back into a stereo track) and then there’s one final bit of jiggery-pokery which is to remove silence gaps. Partly because of latency when using online audio tools like Hangouts, partly because we are both at a certain age, there are occasional pauses in our conversation that break up the flow. Truncating silences (where Audacity looks for quiet points of greater than a certain duration and then chops them out) makes it sound like we’re a bit more quick witted than we actually are. THE MAGIC OF RADIO.
Between sections I bung in a short music clip which helps to provide some audio glue between segments. Again, a little bit from the Caustic track I wrote. Once all of the audio has been edited the final stage in Audacity is to export as an MP3 file.
If you’ve any questions about any of that, feel free to get in touch. I’m also indebted to the wonderful people at From the Rookery End who helped me with some ideas early on in the adventure.