This article is adapted from episode 89 of Current’s podcast The Pub.
I’d say at least once a week somebody asks me: “What’s the best way these days to record an interview with someone who isn’t in the same place as you are?”
There should be a simple, definitive answer to that question by now, considering the state of technology. But there isn’t one.
It’s 2017 and some of us are still shelling out big bucks to put our remote guests into studios where we can connect to them via ISDN or dedicated optical fiber lines. It’s crazy we’re still doing that, right? For public media in particular, this has always struck me as a terrible use of the public’s dime.
It’s crazy that when a studio isn’t an option, we’re still doing tape syncs, or “double-enders” as they’re sometimes called. That’s when you talk to your guest over the phone while someone records his or her end of the conversation in person and then sends you the tape.
Tape syncs have gotten a lot easier in recent years because most people are now walking around with smartphones with which they can record themselves and then instantly email you the files. But I don’t like asking my interviewees to do that, for three reasons:
1) Smartphone mic quality is limited.
2) It’s a pain in the butt for the interviewee, because most smartphones won’t allow you to record on the same device through which you are talking on the phone, which means the interviewee needs two phones.
3) I have editorial concerns about asking my interviewees to record themselves. If I’m interviewing you, I need to feel free to ask you a question so challenging, so offensive, that you might tell me to go **** myself and hang up on me. If I’m relying on you to go through the work of emailing me a big audio file after the interview is over, I am apt to pull my punches.
It is crazy we’re still dealing with these issues. Studio connections are expensive, and tape syncs are laborious at best and journalistically compromising at worst.
Did I mention Skype? Skype often sounds like crap, it’s unreliable, and don’t forget their broadcast terms of service require you to credit them on-air (or on-pod) “at the beginning and end of such use and over the course of the Program in no less than fifteen (15) minute intervals.”
Like they say in the infomercials: THERE’S GOT TO BE A BETTER WAY.
Well, there are other ways, and many of them are better in some respects, but it depends on your situation. I’m now going to review what I think are the three best alternative products on the market right now. I’ve tried them all in real-world interviewing situations many times. None of them is perfect, but each is good for something.
One note before I begin: I am not going to review services that require extra hardware beyond a microphone, such as those offered by Comrex and Tieline. Several fellow radio people have asked me why I don’t use Report-IT, Tieline’s mobile app, which is very popular with NPR and many stations.
What those people usually don’t realize is that those systems require a codec box on the studio end that could run you thousands of dollars. That may indeed be a logical investment for a big enough organization. But in my experience, such services don’t even work much better than the first service I’m going to review, which does not require any box at all, and it’s called …
“ip” stands for “internet protocol,” meaning it works over the internet, and “DTL” stands for “down the line,” as in “I’m talking to her down the line.”
Think of ipDTL as being like Skype, but better-sounding and much more reliable. It’s a live VOIP (voice over IP) connection that works on the regular old public internet, and it’s web-based — it works right in your browser, which means neither you nor your interviewee have to download any special software.
Kevin Leach, the founder of the company that makes ipDTL, is a former BBC engineer. He talked to me via ipDTL from Manchester, England, and I recorded both sides of our conversation with ipDTL’s in-app recording function. We both used high-quality USB mics with our computers and improvised home-studio setups.
Kevin: I’m just building a bit of a fortress here out of duvets and reflection filters and things …
Adam: I’ll go ahead and tent my blanket over my head.
Kevin: Great, great, you’re doing the same thing I am.
Adam: Yep, we’re professionals, Jesus! […] So, the number one reason why this sounds better than a Skype call is that you and I are using good microphones. But that said, even if we used microphones like this with Skype, it would not sound this good — it would not be this stable. What makes what you do different?
Kevin: Several things — the Opus codec, which is utilized by ipDTL, which is in the Chrome web browser. Now, somewhat ironically, Opus was developed partly by Skype initially, but for some reason they’ve never really implemented it properly, so it’s a bit of a lottery when you use Skype as to which codec it’s going to use.
Skype has got all these various bits and pieces going on which are supposed to, in theory, improve the quality — like echo cancellation and noise reduction — because Skype is designed to be used on any hardware.
So let’s say you’ve got a computer with a fan next to the built-in microphone. Skype’s got all sorts of algorithms that try and cut down on that background noise. But of course, there’s only so much you can do, and actually all it really does — all any noise reduction really does — is reduce the overall fidelity of your audio input.
Let’s unpack some of what Leach said. “Codec,” if you don’t know, is short for “coder-decoder.” In this context, it’s a piece of software that takes your raw digital audio and compresses it into a smaller stream of data so that it can move faster and easier over the internet. You can hear that compression on Leach’s voice — he sounds a little digitized. But he still sounds pretty darn good to me, and I can tell you we had minimal delay and ZERO — I repeat, ZERO — dropouts.
As Leach said, ipDTL uses the Opus codec, which apparently works really well. But the other reason his service sounds better than Skype is that it simply eschews all of those fidelity-reducing noise and echo-cancellation features, because Leach assumes if you’re using his service you’re a professional; you know enough to use a good external microphone with your computer and to use headphones so that you don’t get any echo.
The flip side, of course, is that the person you’re interviewing might not be a professional, which leads me to my list of catches for ipDTL.
Catch #1: It’s only as good as the microphones your interviewees have. If they just use their on-board computer mics, odds are it’s going to sound pretty bad.
Catch #2: It’s only as good as your internet connection. I’ve never had a single dropout using ipDTL, even with the dodgy Wi-Fi I have at work, but you could conceivably run into bandwidth problems, especially when you start connecting with multiple guests at once.
Catch #3: The user interface is pretty overwhelming and, I think, a little messy. That doesn’t bother me so much, but there’s been a couple of times when my guests haven’t been able to figure out what to click on to join the call. Leach tells me they have some improvements planned along those lines.
Catch #4: It only works with the Chrome web browser. I should say it works “best” with Chrome; it does also work with browsers like Firefox, Opera and possibly Edge (thought I haven’t tested the latter). What it definitely will not work with is Apple’s browser, Safari, and it’s important to understand why not.
ipDTL, and all of the browser-based services I’m reviewing here, rely on something called WebRTC (Web Real-Time Communication) — that’s an open-source internet protocol developed by Google that, among other things, allows websites like ipDTL’s site to access your device’s microphone and camera. Apple, for reasons known only to them, has thus far refused to get on board with WebRTC, so that’s why ipDTL won’t work on the Safari browser. This leads me to…
Catch #5: ipDTL won’t work on an iPhone. It might work on an Android (though Leach doesn’t recommend it), but it definitely won’t work on an iPhone because iOS uses the Safari browser. True, you can download a version of Chrome for iOS, but that’s basically just Safari with a skin on it that makes it look like Chrome. Safari can’t handle WebRTC, which means ipDTL can’t access the iPhone’s microphone.
This is a big problem for me, because I really want my interviewees to be able to talk to me with their smartphones. It’s convenient for them, and even though iPhone mics aren’t great, at least I know what I’m getting. Computer mics are a lot more unpredictable.
Catch #6: ipDTL’s built-in recorder only records to a single track; you can’t record you and your guests to separate tracks. On the upside, there’s a great built-in mixer so you can balance everything. There’s also a handy tool for playing sound bites down the line. But if you want to record to separate tracks, you’ve got to do that yourself using an external mixing board or some audio routing software, like Loopback, to send the different signals to different tracks on your DAW.
Catch #7: ipDTL is pretty expensive. It’s free for your interviewees, of course, but for you, a day pass is $15. Monthly plans range from $30 to $70. At the higher end you get video, you can connect with up to four people at once, and — get this — you can call ISDNs. ipDTL actually has physical banks of ISDN boxes in Denver and Manchester through which they route your call. It’s pretty cool, but it comes at a cost.
Before I leave ipDTL and move on to the next product I’m going to review, here’s some sound I recorded with ipDTL talking to my friend Ben Harrison, co-host of a Star Trek podcast I like.
(To give you a direct, side-by-side comparison of all three of these services, I wanted to interview the same guy, using the same USB mic with each.)
In this clip, Harrison is complaining with me about ipDTL’s user interface.
Ben: It was confusing to me. I mean, it definitely took me a beat where I was thinking, “Did Adam just send me the wrong link and this is just an advertisement for ipDTL?”
Adam: Yeah, for some reason it just doesn’t look right …
So that’s ipDTL, and again, it sounds pretty good. If you need really high fidelity, however, you might want to look at the second product I’m going to review, which is …
Compared to ipDTL, Zencastr is in a different category of products sometimes referred to as “store and forward” services, but I think of them as “automatic tape syncers.”
Zencastr is another web-based app that works right in your browser — no software to download. Like ipDTL, Zencastr connects you to your guest via a VOIP connection. (In my experience, that connection sounds better than Skype though not as good as ipDTL.)
Here’s the big difference. With Zencastr, that VOIP connection is not the final product — that’s just how you communicate with your guest. While you do the interview, Zencastr simultaneously records both sides of your conversation to your respective local drives. By that I mean, it records the uncompressed, high-fidelity sound of your voices to the hard drives on your computers, and then when the interview is over, Zencastr automatically uploads those files to your Dropbox account, and you can download either the combined product or the isolated tracks.
To demonstrate how this works, here’s a clip of Harrison (complaining about one thing he doesn’t like about the Zencastr interface) recorded two different ways. In this first version, you’ll hear Harrison via the VOIP connection that I heard while talking to him over Zencastr. In the second, you’ll hear the uncompressed WAV file that Zencastr recorded to Harrison’s local drive and then automatically uploaded after the interview was over.
One thing I will say about this — I think I’m fairly savvy with these kinds of things, but the way to change what microphone was feeding into this site is just an icon of a speaker inside of a gear, and I don’t think that my grandma would know what that was.
Pretty great sound in that second version, right? What’s more, there will never, ever be a dropout when using Zencastr, because the sound is being recorded before it goes over the internet.
The obvious downside of these automatic tape sync services is that after the interview is over, you and your guest have to wait for those big WAV files to upload. If I had asked Harrison a question that really pissed him off, maybe he wouldn’t have been inclined to wait. Or, what if we simply got cut off before the files finished uploading?
This leads me to my favorite thing about Zencastr — it actually records two files to your respective local drives: the uncompressed WAV file, and a compressed MP3 file. Here’s Zencastr founder Josh Nielsen explaining why this is advantageous in an interview I recorded with Zencastr.
Zencastr grabs the audio stream from your microphone and then it saves it in its raw form, and then it also encodes it to an MP3, and then it uploads that MP3 to Dropbox. And this is all happening in the browser while we’re talking. And then at the end, the MP3 is already uploaded, so it usually finishes right away. And then the WAV uploads kick in right after that.
Get it? The MP3 version starts uploading during the interview, so it’s going to be there for you even if you get cut off, which is exactly what happened when I was interviewing Nielsen. He lives in a little village in Thailand, and his internet is very spotty — his connection died in the middle of that interview, so I never got his WAV file, but I did get his MP3, which is the sound you just heard. It sounds a little compressed, but still very good — better than ipDTL, for sure.
While Zencastr is not the remote interview solution we’ve all been waiting for, it’s probably my favorite of the bunch. Another thing I like about it is it has a menu where you can load at least five sound bites in advance that you can play down the line (ipDTL only lets you load one at a time).
One feature I would never use but that could be incredibly useful to amateur podcasters is automatic postproduction; Zencastr can mix all your tracks together, balance all the levels, and then master the file to your desired loudness. I want to control those things myself, but for the amateur user that could be great.
Lastly, Zencastr is pretty cheap. The pro plan is $20 a month, or there’s a free plan with limited features that maxes out at eight hours of recording per month. Of course, it’s always free for your guests.
Catches for Zencastr:
Catch #1: As I mentioned, Zencastr works in coordination with Dropbox, so you’ve got to have a Dropbox account too. That’s not a big deal; you can get one for free, but if it starts to fill up you’ve got to pay for extra storage.
Catch #2: As with ipDTL, Zencastr is only as good as the hardware you and your guests are using.
Catch #3: As with ipDTL, it’s only as good as the internet you’re using, and I’ve had trouble when interviewing more than two people at once. However, even if the VOIP connection starts to sound ratty, it doesn’t matter that much, because that’s not the audio you’re actually going to use in your finished product.
Catch #4: Unlike ipDTL, you can’t use Zencastr live. Well, I suppose you could, but it wouldn’t sound very good. You’ve got to wait for those locally recorded files to upload after the interview is over to get the full effect.
Catch #5: As with ipDTL, Zencastr works best with Chrome and won’t work at all in Safari, because it relies on the WebRTC protocol that Apple has thus far refused to integrate.
Catch #6: As with ipDTL, Zencastr doesn’t work on iPhones, for the same reason.
However, the third and final product I’m going to review will work on iPhones, and it’s called …
Ringr is another automatic tape sync service. Like Zencastr, there’s a web-based version you can use right in your browser, but unlike Zencastr, Ringr also has a standalone mobile app. That means your guests can talk to you with their iPhones; they have to download the app, which is a pain, but at least it’s free.
Here’s two versions of a sound bite from my 2015 interview with Ringr founder Tim Sinclair. At the time, Ringr only had the mobile app, not the web-based version, so both of us used the mobile app on our iPhones while I simultaneously recorded myself through my normal hosting mic onto my computer — that’s the audio of me you’re going to hear.
In the first version, you’ll hear Sinclair as I heard him during the interview via Ringr’s VOIP connection, and in the second you’ll hear the WAV file that Ringr automatically recorded on Sinclair’s iPhone and uploaded to me after the interview.
Adam: Hi Tim, how are you?
Tim: I’m fantastic, Adam, how are you?
Adam: Let me ask you the classic radio-guy levels question: What did you have for breakfast?
Tim: [laughs] I had a Clif Bar and a Diet Dr Pepper.
Even the WAV version doesn’t sound great; that’s because he was talking into an iPhone mic, and iPhone mics just aren’t that great. But they’re pretty good, and you know what you’re getting, unlike a computer mic.
So, if I’m going to interview someone who has an iPhone but doesn’t have a good USB computer mic, I’m going to reach for Ringr instead of Zencastr. It’s also nice that Ringr handles all of your data storage for you; no need to use Dropbox, like with Zencastr.
Ringr’s user interface is decent and the pricing is reasonable. A basic plan is $8 per month, but that only gets you an MP3 that combines both sides of the conversation. If you want WAV files and isolated tracks, it’s $19 per month, which is still reasonable.
Catches for Ringr:
Catch #1: There’s no free option or day pass, so it’s really not good for sporadic use. The least you can buy is a whole month.
Catch #2: As of now, you can only record with one guest at a time. Sinclair has been talking for more than a year about adding conference calling, but it hasn’t happened yet.
Catch #3: I have found the VOIP connections to be very unreliable in the past, though they seem to have improved lately.
Catch #4: Like Ringr and Zencastr, you can’t use the web-based version with Safari.
Catch #5: Unlike Zencastr, Ringr does not upload an MP3 version while you’re doing the interview. It waits until you terminate the connection to start uploading, which means you just have to trust that your guest will leave the app open while the files upload.
Catch #6: If you or your guest is using the mobile app, you have to remember to put the phone on Do Not Disturb mode because incoming phone calls will disconnect your Ringr conversation.
Further complicating matters is the fact that the iPhone’s basic Do Not Disturb mode only blocks calls while the phone is locked, and your phone obviously won’t be locked while you’re using Ringr. Also, the basic Do Not Disturb allows calls from people on your “favorites” list. You have to go into the Do Not Disturb settings and turn both of those defaults off.
This is a huge pain, especially if you have to explain it all to you interviewee. Honestly, it might be easier to just get disconnected.
Catch #7: If you, the interviewer, want to use the Ringr iPhone app to initiate the connection, the app sends you to a virtual holding pen while it waits for your guest to join the conversation. While you’re waiting, you can’t do anything else on your phone; your screen is totally filled with this Ringr waiting message.
Now, what if you’re waiting and the guest doesn’t show up? You’re going to start worrying that maybe something is wrong; you’re going to want to text or email or call your guest to find out what’s up, but in order to do that you have to leave the app, and what if the guest tries to join just as you’ve left?
Here’s what I do instead. As the interviewer, I don’t ever use the mobile app. It’s fine if my guest uses the mobile app, but I only use the web-based version on my computer, where the holding pen is only a pop-out window that doesn’t monopolize the whole machine. That way I can still check my email, text or make a phone call while I’m waiting for my guest to join, and as a bonus I can record myself right into Ringr using my USB mic.
So, which one should you use?
I’d say use ipDTL if you can afford it, if your interviewees have good computer mics, and if you don’t need super high-fidelity. One more time, here’s Ben Harrison using his USB mic on ipDTL …
I would say use Zencastr if your interviewees have a good computer mic, and if you do need super high-fidelity. One more time, Harrison with his USB mic using Zencastr …
Lastly, I would use Ringr if your interviewee needs to use an iPhone instead of an Android or a computer mic. Here’s Harrison on his iPhone talking to me with Ringr …
I will say that I’m getting more plosives from you than I’m used to hearing on a phone call, and I’m wondering if that’s how you’re holding your phone …
(Aside to Harrison: No, my phone-holding technique is impeccable, and stop talking about my plosives.)
So here we are, 4,000 words into this article, and that’s the best conclusion I can give you. This is why when people ask me, “Hey, what’s the best way to record a remote interview these days,” I usually just find a way to change the subject.
Adam Ragusea hosts Current’s weekly podcast The Pub and is a journalist in residence and visiting assistant professor at Mercer University’s Center for Collaborative Journalism.
Yo Adam – worth noting:
Apple sort-of added WebRTC support to the iOS safari dev kit but that was nearly a year ago and there’s been nada since then. I get the feeling Apple iOS will natively support WebRTC over Tim Cook’s cold, dead body at this point.
Yeah, I saw that a year ago and got really excited. And then…crickets. I’ve given up. If they’re developing their own thing, that’s great, but they’re way behind the pack.
I think there’s still a lot of market forces pushing against Apple to ever natively support WebRTC. Not just their own (worthless) Facetime system, but also various politics tied up in how wireless carriers provision data for different purposes.
Unless and until you can create a simple webpage that has a simple list of links that anyone in an iPhone can tap the link and it will automatically “call” a SIP number? Then the true promise of WebRTC will never be realized.
All of these technologies are interesting but the caveats are ridiculous. That’s not your fault, of course, Adam. It’s just outrageous that we still have to have all these caveats in 2017. As a podcast host setting up specific interviews here and there, I think you can get away with it for most of your interview needs. But all of these things are functionally worthless for most news radio stations. There’s no way you’re ever going to convince a random newsmaker to jump through all these technical hoops just to make a “call”. Half the time it’s a battle to get them to stop using the speakerphone, never mind to not use a cellphone, when you’re calling them.
Skype is, hands-down, far more accessible since a lot of people know what it is and might even have an account and the app on their smartphone already. But as you say, it’s prone to dropouts and lousy audio quality, and the TOS is ridiculous.
Yeah, I think they’re all transitional technologies until we get to near-universal VoLTE / HD voice on everybody’s phones.
Great and informative review, Adam. I really appreciated your thorough approach. Zencastr seems to have the most balanced approach (and the best audio). It’s amazing how much technology has improved in this area, when the best you could hope for was a Comrex, phone hybrid or ISDN solution. This really flattens the curve, and makes a long-distance guest sound almost as good as if they’re in the studio with you.
Great review. Thanks for taking the time to break this down.
On a related note, Adam – one thing that also intrigues me is the idea of a cloud-based multitrack audio editor that runs in the browser. So far there are some options in this realm but they’re all kinda iffy.
But today’s news about Firefox could very well push a lot more development in this arena.
So far audio/video editing is one of the biggest remaining reasons why one can’t just use, say, Chromebooks for all your station’s staff computing needs. Right now the G Suite for Business and Chromebook approach has largely taken over the K-12 public education realm, and it’s making huge inroads to the higher ed space, too.
And on Android & Windows Phone or tablet or others? Well, it’s very interesting. And we want a new tool to make a podcast, but did you know? Patagonia Clothing Company is a new underwriter for the PBS NewsHour. #PBSNEWS
Thank you for this! I’ve been doing Skype recordings, which were easy for the guest, but sound less than optimal. And, considering who my guests usually are (elected officials or other politicians, often in a hurry), I really didn’t feel I could even consider asking them to record their end and email it to me. Doing a phone interview? Always up for that.
So, I’ve been intrigued with these services. I found your article through a search for “Zencastr review,” and I’m glad I did. The problems with iPhone and Safari are a kick in the teeth, though, as almost all my guests will want to do their side through their phone, and many of them are on iPhones. So, I’ll probably go with Ringr for the iPhone folks, and Zencastr for the others. (Have to think about cost, though.)
Again, thanks for this excellent and informative review.
Would love to read about how https://tryca.st/ compares here
The new version of Skype for Mac has no volume control. How are other podcasters dealing with this?
How are all of you recording Skype calls? Also, has anyone considered doing a video call with software like BeLive or ECamm Live, and then stripping the audio off of that? That way you have a video for YouTube and audio for a podcast.
Has anyone tried Cleanfeed.net ? This seems to be a very simple system developed by UK broadcasters and it uses the Opus codec. There is a free version with no strings and no ads.
It uses Chrome and you just log in, send an email to your remote reporters computer or smartphone and it just works. You can also record the 2 ends of the conversation directly on separate tracks.
The only problem I have is that I can’t get my usb inbterfaced professional mic to be recognised on my Android Smartphone within Cleanfeed (set to USAB audio for the phone) – although the phone will record from it eg to the Field Recorder app. (and the mic will work on Cleanfeed at the ‘studio’ laptop end.
Any ideas how this might be solved ?
Great work, Adam, and now in these viral times your article becomes even more relevant. Took your work and expanded on it a bit: https://hearingvoices.com/2020/03/remote-recording/