The fourth part in a series on making your own podcast that won’t break the bank, concentrating on DIY and low-cost podcasting. Read part one, part two and part three first.
Plug in babies
If your mic is USB – and I really recommend that for first timers – you can skip to the next section (bonus points if you do a moonwalk slide).
But say you find a mic at the back of a cupboard, or borrow one from friends and it’s the older type? You need to plug your mic into something to turn those sounds in the children’s scribble that is called a waveform. What you use depends on how much you want to spend, if you’re going to have guests or cohosts, what kind of show you’re going to record.
The first type is an audio interface. previously a sound card but thr move to modular USB devices and miniaturisation usually means the device is a external. This could be anything from a cheap USB dongle, to a Mic?Line in already on your computer to a multi-input behemoth. The minimum is at least one input (usually expressed as 1×1, 1×2, etc). This can be a 1/8″ 3.5mm jack, a 1/4″ 6.5mm jack or an XLR input. XLR is an an audio connector beloved of studios and live performance spaces, and looks like a surprised face, 3 dots in a circle. You can get cables tor adapters that convert one type to another.
As with anything, more is better…whatever you can afford. The better interfaces have quieter pre-amplifier – the bit that makes the microphone louder, and more options. A headphone monitor output with a volume control is really useful. If you have guests then another input is useful, but check if your device and software can cope with recording multiple inputs at once. Most can, but phones and some older computers will struggle or not support that.
Brands that are good: In the past I’ve used M-Audio and Native Instruments, those are good as well as Roland (see the note below though). My next purchase if I get one will be Focusrite Scarlett, but again those aren’t exactly cheap. Tascam makes good gear, as does Steinberg but seems their low-end devices have been upgraded as has unfortunately the price. Not tried the Behringer interfaces, make sure they have the right inputs/connectors you need. Unlike mixers, soundcards don’t have many moving parts, so second-hand is a good possibility – check that the interface is still supported by your device though, you can get drivers and support by going to the manufacturer’s website.
One of the reasons I’m less happy about M-Audio these days is they dumped support for Firewire – avoid interfaces which use that because they probably won’t have modern drivers, so it stopped working when I upgraded my computer. And Native Instruments has done likewise recent the Audio4DJ, it doesn’t work with OSX High Sierra. Irritating because your kettle or fridge doesn’t stop working when the manufacturer brings out a new model, why should they? Grr.
The other type of device for getting sound into your device is mixers, it used to be that you had to get an interface AND a mixer, but modern mixers have a USB interface built in. Why a mixer? Well mixers are needed if you are combining more than one audio signal together into one stereo output – say for a DJ set or a podcast, the theory is the same.
So if you have more than one microphone, a music show especially with external audio sources like vinyl or an iPod, a guest on Skype or really advanced a cohosted music show with one of the people on Skype – you’ll need to mix those audio sources somehow. Enter the mixer.
Like the interface you probably do need to future proof yourself by getting something with more channels than what you need unless you are 100% certain you’ll never use them. Or it’s the best you can afford.
Although it’s possible to do these things purely inside your computer (see the next section), it’s a lot easier to route audio physically with cables, you can usually see if the audio signal is there visually with LEDs and troubleshoot issues. If you do want to do Skype recording, you might need a slightly more advanced feature called ‘Sends’ also known as FX sends, so the person can hear you and/or the music you are playing, and/or multiple inputs on your interface. It depends on how the USB mixer feature is implemented.
So for a mixer I’d avoid anything less than 3 channels – because in mixer-land each channel is usually mono, with usually the ability to mix in one or more stereo inputs from ‘tape’ or ‘CD’ without fader, EQ or any controls. So for a basic mixer 3-4 channels is the starting point, less than that you may as well use an interface because you’re not really mixing, and won’t be that future proof.
Interface or mixer, which one you use depends in part what you’re trying to do. If you’re recording a solo podcast with no external music sources, then go for a simple interface. If you’re recording more than one voice, with possible external inputs, say from CD, vinyl, etc. then a mixer is probably the way to go.
Brands for mixers that are good: Mackie is great, amazingly quiet pre-amps (pre-amplifier is the thing that brings microphones up to a ‘line’ level), but expensive. Allen & Heath and Peavey are great but are also similarly priced. Mid to low-range is Alesis, Numark and Yamaha which are also reliable. Lower down in the stack are the likes of Behringer which personally I found rather noisy, but the non-USB version of Xenyx was my first proper mixer (well my first was an old audio mixer for video, but that lacked EQ). Pyle as well are apparently well regarded but I’ve not tried them…seems hard to get in the UK? As said before, get the best you can afford.
I’d avoid second-hand mixers unless they have been expertly refurbished, especially from music studios or those than smoke or live in dusty/dirty environments. Mixers have moving parts and these can easily get worn out, get trashed or covered in muck. If it’s from a friend and you know how well it’s been treated, or it’s one of the more high-end makes (Mackie et al) which are designed to be used heavily and last a long time and might be OK – go ahead. Otherwise it’s probably a false economy.
Virtual Insanity (without a silly hat)
So what if youve not got the space, or use a USB mic but have a friend coming around with another mic, or want to do a music show? Enter virtual mixers and virtual audio routing. You can combine, re-route, send and mix audio without leaving your computer. Thing is, it’s a lot more fiddly than a physical mixer unless you get some software that makes it simpler. Usually these works as a virtual soundcard driver, which rather than sending the audio to your speaker then sends it to other programs, to a recording input in your DAW, to Skype – whatever you want.
If you are on Windows – pat yourself on the back cos this is one of the areas that Windoes pisses all over OSX, annoyingly (said the Mac user). You can get a free program called Voicemeeter Banana and it will work just like a mixer, so you can route and combine multiple mics, sources. Etc. My friend Ian over at the excellent Lloydbrary podcast uses Voicemeeter and it works perfectly. I’m jealous! There are quite a few other virtual soundcard devices – I used to use TotalRecorder in my pre-Mac days, if that’s still going? – which are free or cheap.
If you are on a Mac, prepare to either spend some money or fiddle with settings and/or open source software. First port of call it to experiment with what Apple calls ‘Aggregate Devices’, because for the majority of simple setups it’ll probably suit you best and it’s free!
These mean you can create your own virtual soundcard, adding inputs and outputs to one single device. Some have used this to cohost with Skype, I couldn’t get it to work – but this is how you can use one input from a different device or card and another output at the same time, or if your software like Traktor allows multiple outs, you can route your monitor mix to your speakers while outputting your master mix to a virtual output to your recording software. It won’t mix two or more sources together though, for that you’ll need other software.
I don’t like Rogue Amoeba’s rather expensive Loopback, I mean $99 for something you can get free on Windows? There is also rival Sound Siphon at a cheaper $39 but really this should be free or built in…Not sure SS does the same as Loopback, as ever download the trials and fiddle with them.
I do have a cheeky free alternative, download one of the screen capture/recorders from Apowersoft – I forget which one, I think it was either Screen Capture or Screen Recorder, and it’ll install a handy virtual device for you to use. Which although I uninstalled the app ages ago, it left the device there and it’s not time-limited. This is what I use to get my DJ audio into REAPER.
There is a few open source alternatives as well – JACK and Soundflower which if you are on Linux the former should also work well for you. I had some issues with it, because both my DJ programs, Traktor and Serato flatly refused to open when I installed JACK. Annoying cos it otherwise did exactly what I wanted to do, for free! And Soundflower works, but I’ve always found the signal very quiet.
If you just want to record Skype, you might find Audio Hijack combined with one of these virtual routing systems or Aggregate Device is a cheaper way to go. That’s the way I’ll go if I want to start cohosting on Skype again and it ‘only’ costs $59. Hmmph. RA, you really need to sort out your pricing, you’ve always been expensive! Which brings me to:
Recording podcasts with a remote host or interview guest especially a music show where the host needs to hear the music is way beyond this guide – think of that as super-advanced, the End of Level Big Boss of podcasting – it’s still something that still gives me nightmares to this day. You could ask the person to record their side on their device and send it to you, which is fine for a regular co-host or a fellow podcaster, but not so great for guests who wouldn’t even know which end of the mic to talk into. And you are one crash away from losing the whole show – it’s best to record both the Skype or VOIP stream AND individual parts because of this. Be paranoid is my motto, well about data loss anyway!
When recording remotely, hardware makes it a lot easier especially if you have a mixer that can do channel effect sends (so your host can hear you and/or the music!) or interface or USB mixer with more than 2 inputs and outputs and quite a few cables. But if you’re in the virtual domain, you might want to avoid using Skype as the quality can be frankly pants, and you can’t record each host separately. This is fairly essential as editing a natural conversation where people talk over each other, usually at the exact the point you need, well it’s a nightmare.
I am planning to get back into having remote podcasting hosts (when I did it before I used my big mixer and interface) and if I do I’ll probably use one of the VOIP recording/meeting hosting services. Think ‘Hangouts’ with knobs on, with recording and integration built in. Popular ones for podcasters are Zoom which has a free level at upto 40 mins and is more of a business meeting system but a lot of ‘pro’-casters use it. Zencastr has a free tier for upto 8 hours which looks the best for me and automates the remote recording I mentioned earlier. Others which cost or only have a free trial are Cast which also can act as a podcasting host (more about them in the next part), ipDTL, Ringr which works with phones, etc.
Apart from Ringr’s app they tend to work from within a browser, usually Chrome or Firefox (Apple doesn’t allow microphones to be used directly in Safari, being the usual killjoy it is renowned for). I haven’t yet worked out how you allow someone to hear playing music especially as I think browsers only allow one input and output…maybe I could use Audio Hijack, or use the internal soundboard that some of these services have and upload all the tracks.
As I said, it’s a complete headache and I’d really not suggest a beginning podcaster to do it unless they are doing a simple talking-head show, and even then….start simple.
Something to watch for with virtual mixing is something I’ve mentioned before, latency. This is the delay between the signal entering the computer, or from where it’s been routed, to where it ends up. This can make monitoring a pain as you get an echo. There are settings to reduce it, but you will never get rid of it especially listening back ‘monitoring’ your own vocals live – the computer or device isn’t a time traveller, there will always be a delay.
Why listen to yourself while you record? Well, it’s hard to explain, it seems that humans have this need to have a return from what sound they create to their ears – this is what makes the anechoic chamber so spooky, that the echo doesn’t come back. Part of this feedback loop helps you to check how you are being recorded, if you’re being too loud, too soft, etc. The best place to monitor that is the mixer, not the computer or recording device, or on the mic itself if it’s a USB one, because there is usually no or little delay/latency. If it’s an analog system, usually the delay is tiny, I used to monitor via my Sony PCM or Minidisc which was last in the chain…but do that with a digital recording and you’ll probably sound like you’re on the moon.
But if you can’t afford that, you might have to record without monitoring and listen back, and it’s just a little slower, that’s all. But you still need something to listen on, which brings me to…
Headphones come in all shapes and sizes, but again it’s another area where people start to become hi-fi nerds. Yes, Beyer DT990’s are amazing, but the headphones I’ve used for recording, DJing and the like have been rather more prosaic. Earbuds are more than fine for podcasting, they’re portable, and pretty durable and you tend to have them anyway. Working up from that, any decentish make, but avoid what are called ‘open-back’ headphones. I’ve made that mistake, basically means they will feedback next to a mic when recording, which as well as painful is rather an annoying mistake. Go for closed-back or in-ear phones.
I recommend Sennheiser, as they tend to be fairly flat/neutral sounding. I have a battered pair of HD 25 SP’s which you saw in the first part, which are like the proverbial spade, most of it has been replaced, but they soldier on. AKG, Beyer, Audio-Technica – you’d probably not go wrong with those but tend on the expensive side, but not all, the cheaper versions are usually good value.
Phillips are surprisingly good, as well as Sony, Apple’s own iPod earbuds are fine as well. Skullcandy are a brand that I’ve tried to hate, all ‘urban’ and wacky but unlike Beats et al, they are also not bad although I’d not use them for mastering like Sony headphones tend to be rather bass heavy. And also no-brand and less high-street brands can be good as well. This market moves so quickly, I’d check reviews on WhatHiFi, Which, Amazon and other sites, cross reference them, use Fakespot – be aware of biased reviews, or lemons, but if several well-known sites say a pair is good, then it’s probably correct.
I’d avoid using Bluetooth phones though, unless you use the rare/higher-end aptx codec, it compresses the sound and quite often does funny things with DAWs. I tried it with REAPER and whereas iTunes et al are fine it suddenly goes a bit crazy/distorted and I have to switch to a wired cable. Also BT connections can drop, they aren’t 100%, which of course you’d not want in the middle of an important interview in a quiet space for the connection to drop and suddenly blare out into the cafe or train or wherever.
Obviously if you have a high-end headset, earPods etc you might be fine, but be aware that audio quality can suffer as they aren’t made for podcasting but for squeezing down telephone wires and audibility, not hi-fidelity. And Bluetooth connections like market shares can go down as well as up!
So, you’ve recorded your first podcast be it simply via Sound Recorder, your laptop mic, a USB mic and a DAW, or your phone…(yes I can’t show you how to do that, there are too many ways and too many pieces of software out there. Manuals, help sites, forums and even courses in your software can help you there).
So what to export as? MP3. I don’t care what anyone else says, the only choice is MP3. It might not be the ‘best’ format, in fact it’s rather old in the tooth now and I personally use AAC/M4A for the music I rip or convert. But MP3 is the most supported across websites, podcast software, iTunes, Google Play, Spotify if you’re lucky enough to get on there, iPods and other MP3 players, desktop software…it might not have handy ‘chapters’ like AAC and whatever that odd Audible format is, but using those will shut out a lot of Linux and Windows users, and those with older devices that don’t support MPEG4 aka M4A/AAC.
Why? Well AAC is licensed, whereas MP3 has expired or been retroactively engineered, so it basically is license-free now. So it doesn’t cost people to add it to their hardware or software.
So the settings – this is a bit techy and there is no getting around that, it’s all a bit nerdy by nature so Google anything you don’t know:
Variable bitrate (VBR) is a good thing to enable, so set the target bitrate (bitrate is how much you’re compressing the file, higher the number is better, but a bigger file size) at roughly your target bitrate and it adjusts the bitrate automatically reduces the size of the show.
Not many reasons to change the sample rate nowadays (this is the range of frequencies encoded, for reasons too nerdy to mention it’s double the frequencies you might capture), and in fact like mono it can introduce all kinds of issues with playback so unless you suddenly want to sound like a Chipmunk, stay at 44khz or 48khz.
As regards to bitrate settings, it depends on what your show is. Spoken word only shows can go as low as 64kbs bitrate, but I’d bump that up unless you’re short of space or know you’re aiming for mobile listeners or those without broadband connection. Ideally more closer to 96kbs or 128kbs, I personally wouldn’t go below 128kbs nowadays cos you might get the dreaded artifacts, the watery, whispery sounds. Music shows need higher bitrates, again no lower than 128kbs unless there is a good reason. My shows are at 70-80% VBR which work out at around 200-250kbs.
If you’re using Constant Bit Rate, I’d bump those up a bit too, cos VBR works at a range between X and Y, and averages out to create the target bitrate, and thus contains some higher and some lower, whereas CBR is only that, no higher.
Stereo. Unless you are spoken word, always stereo. Avoid Joint Stereo, Intensity Stereo, M/S Stereo, L+R stereo, all those ‘clever’ things meant to reduce the size ‘psychoacoustically’. Nope, generally they do funny things when they get confused with certain tracks and can cause issues with playback. Although I’m sure now I’ll get the nerds at HydrogenAudio coming at me with knives, I’ve found they create weird effects especially at lower bitrates (128-192 or below). For the frankly small saving in file size you’ll make, it won’t be worth the risk of odd whispery noises all over your podcast.
Mono is common in spoken word, I have mixed feelings about that, I have come across a few players that don’t understand mono MP3s or play nor convert them properly. One odd thing is my live-streaming radio station (Classic Bootleg Radio) flatly refuses to work with mono files, which was a problem when I was re-broadcasting friend’s podcasts. I had to laboriously convert them all to dual channel stereo. So I personally wouldn’t, but if upload/webspace is limited?
As with anything, try different rates and see what the file sizes are, test and listen on various devices, phones, websites, iPads, iPods, MP3 players, anything you can get your hands on – especially listening on a laptop or cheap speaker cos they quite often reveal stereo > mono issues. I find something that starts to just sound bad on my super-pro Beyer headphones is best, because that means all things below that it’ll sound good. It’s a trade off. I found some odd artifacts with my older shows revealed in almost-mono speaker type settings, caused by bass. Filtering the EQ helped that, and switching off that Joint Stereo crap (grr).
Coming up in part five: Domains and websites, blogs and themes, submission that doesn’t involve whips or chains, social meeja and how to make friends and influence people, well at least get them to listen to your podcast. Maybe. Which I guess begging is a different form of submission?