Robin Leboe is on a mission to virtualize the recording studio with Sessionwire, a suite of tools enabling professional and amateur musicians around the world to collaborate remotely. The JUCE team interviews Robin in Sessionwire itself, via its integrated studio video chat.
How are you?
I’m good. Just another crazy day in Sessionwire paradise…
You’ve built a whole empire in this app! You studied guitar, became an audio engineer, worked in studios, worked on the RADAR digital multitrack recorders, taught audio engineering and interactive audio… Sessionwire seems to perfectly combine all of your experience.
One of our main users, Glenn Rosenstein, co-owner of Fame Studios, is a huge fan. He literally said he couldn’t have the business he has without Sessionwire.
We were talking about that and we both agree — there’s a Canadian expression, you know — this is where’s the puck’s going. The industry is moving in a direction where people will be forced by certain factors to work remotely. Financial factors, the budgets are low (or non-existent), the need to reach people outside of your geographic area, etc. There’s financial reasons, geographic reasons, speed reasons.
Using Sessionwire to get stuff done is a thousand times faster than “Oh, I’ll email you a file and you can listen to it, write me notes.” There’s this slack in the chain in that process.
When I started Sessionwire, I used to say “I want to bring back the happy accidents.” If you’re working with somebody and they’re messing around, and all of a sudden they hit this chord and you go, “What did you just do?” You get the goosebumps. You don’t get that in the other remote work framework — the guy goes “Oh crap, I played the wrong chord” and then goes back and plays what you wanted him to play.
In a way, you are aiming to fulfill the internet’s basic promise to musicians. We could just use zoom, but the audio quality doesn’t work for music. I can send you tracks, but then it’s async, not real time. You are collapsing those varied friction points down to a single point.
You’re right on the money. We finally came to the conclusion that Sessionwire is a virtual studio. People that are on here every day say “This is my studio. This is where I do my work.”
So it’s a non-physical representation of a physical thing, a studio. Being in the same room or in the same studio together is the best things, hands down. I want Sessionwire to be the next best thing to that.
We just added cloud storage. I had people saying “Oh, you need to just focus on video.” But if I’m in the studio, I say, “Hey, can you pass me that USB drive?” or go throw tape on the two inch machine. Now I have a cloud storage here where I can have files, copy a link, past it in studio chat. It closes part of the gap of the transfer of information that’s not video and not “liquid” audio.
It’s another plank in that foundation of “This is going to be just like a recording studio.”
It’s interesting that you aim to be one-to-one with a physical studio. Building from first principles, not just adding features because of customer or competitor pressure.
We do spend a lot of time talking to our members. For instance, there was a lot of pressure on us to do multi-channel audio. Something in me was not quite vibing on that.
One guy in Las Vegas does a ton of Atmos mixes. He said “Don’t waste your time on that, at least not right now. I just sent an Atmos mix to Los Angeles a couple of weeks ago. I went there and the room was so bad, I had to bring in somebody from Dolby to recalibrate the room. My Atmos mix is not going to necessarily translate, especially if you’re talking about mid or lower level studios. You’re way better off sending a binaural mix over Sessionwire over headphones.”
But there are little things, like the color of the lights on the side. We had somebody who was colorblind who said he couldn’t see when the light changes, so we changed the color and they’re like, “Wow, I love you guys.”
The customers who say they wouldn’t have their job without you — does that keep you up at night? It’s the best type of feedback, but quite the responsibility…
Yeah, the only thing that keeps me up at night sometimes is the responsibility of what we’ve created — the reliance that people are coming to have on the platform.
Obviously, we are an internet-based product and sometimes things go bad on the internet.
And we have technologies across so many domains. We have the JUCE plugins on Windows and Mac, three formats on Mac, two on Windows. We have virtual drivers on Mac. We have a virtual driver on Windows (we license a third-party product for that, to be replaced one day). And then we have all of the web-based stuff, the video and the talkback and separate stereo audio streaming. We also have cloud storage with two separate cloud storage providers.
So we’re in web audio, we’re into web RTC, now iOS. We just have a new app I finished called Patch on the Mac, which allows you to select any white-listed application and write the audio to our high-quality audio channel. Some of the big record labels were asking for the ability, some A&R person wants to play a YouTube video or Apple Music or Spotify or whatever.
All these different domains. So sometimes I wake up at three o’clock in the morning and go [sharp inhale].
The surface area seems daunting. A lot of the job must be maintenance? Software crumbles unless coddled. What does your team look like?
Very small team. There’s myself and John Wilson, our senior dev. And Steve Yee. They do pair programming extraordinarily well. My job has become more of the architect. I’m kind of leading the charge on the macOS and iOS stuff right now.
Wow. So small!
Thank you for saying that. I’ve talked to investors about our team over the years and I go, “You have no idea what a gem you have here”. The amount of stuff we’ve been able to get done is unbelievable.
Who worked on your JUCE plugins? Was that outsourced or in-house?
Originally it was outsourced and we’ve been in the process of bringing it in-house again.
When Covid hit, we went from being background noise to getting a hundred emails a day: “Can our choir rehearse on Sessionwire?” All this kind of stuff. At the time we were still an A-to-B type product where it was just person A and person B.
And Zoom was there, where you can have a whole bunch of people. We had to quickly redesign the entire platform from the ground-up. We got it re-engineered so that we could have many people on a call. At that point, we did not have plugins. And everyone was going “Oh, you need a plugin!”. So we brought in a third-party.
I looked through your CV and other interviews you’ve given. I had you pegged as a music recording guy, but when did you get into dev? What did that journey look like?
I started around probably 1996 or 1998, something like that. I started doing a bunch of Visual Basic stuff. I built a time tracking application for a studio that was doing a bunch of film stuff. And then I got into web stuff. I ended up being a sole proprietor and built out e-commerce stores. So I did a lot of work in that that space and it kinda went from there.
I don’t enjoy dev where it’s like “I’m the dev, give me something to do and I do it.” I’m not that guy. It’s more like “What if we could do this? Oh, you’re busy? Ok, I’m diving in.” So, I just built this menu bar app and it works great, so I guess that makes me a dev?
I don’t want to do what what John and Steve are doing. We’ve got like millions of lines of code and a big monorepo. I just get tired even looking at. There’s so much going on in there and all the devops stuff and all that stuff is cumulatively insane.
You mentioned Sessionwire was Mac-only until the pandemic, when the JUCE plugins came out? It sounds like the pandemic really impacted the business.
Absolutely. It compressed ten years down into like three in terms of adoption.
Yeah — or 3 months! So you also lead the technical side of the operation when it comes to the desktop apps?
Yeah, I have a lot of ideas around that. When we first started Sessionwire in 2016, we had a full-on MIDI pipe with MIDI timecode, MIDI machine control. I could hit a button on my computer and start your DAW playing. We had timecode burn-in, the whole thing. We were targeting the high-level producers.
But what we encountered is that a lot of high-level people, even quite technically skilled in audio-related stuff, have a real mental block about computer technology. So we’re like “Oh, you just have to setup this IAC driver and blah, blah, blah” and they’re just like, “What?”
The pro audio industry can lag in terms of technology adoption. Nobody wants to upgrade. Nobody wants OS updates. In order to do “stable business,” people become conservative in their digital technology choices. At least that’s how it was.
It’s interesting to watch adoption curves, I’m really grateful I went through a couple of them. My first synthesizer was a Prophet 5 and it had an RS-232 port on it. It did not have MIDI. I remember when the DX7 came out. And then the Alesis MIDIVerb came out…
Keep talking to me about synthesizers! [laugh]
Talk about a workhorse, MIDI is still in use every single day all over the place and it was such a simple slow thing and it just did the job. And now of course they have MIDI 2.0—
Well, not quite yet! It’s taking time…
I know. Again, it’s an adoption thing and they’re pushing hard. A lot of people are like, “If it ain’t broke, don’t fix it.”
Yeah, the question is if only having 128 values is “broken!” And then there’s MPE…
I know, I know. But for your average bear, you plug it in, and hit the key and it works.
But adoption curves — a two-inch machine worked great. People loved it. I love the smell of a new box of Ampex tape. Put the thing on, it had a great vibe. You got all the meters and producers with big chains and the big mustaches vibe, right?
It took a long time — we jumped in with Radar to say, “Hey, you take that tape machine, put it in a box this big, stick it in a rack, you still got the auto locator.” We eventually sold a lot of them. But then we got leapfrogged by the DAW, which I saw coming so far before my boss at the time…
It took some professionals many years to finally go digital. Now, enough time has passed where there’s young people who are going back, rejecting the computer, romanticizing an era they weren’t even alive for…
I used to teach tape editing with a razor blade and tape. It’s valuable and I’m so glad that I and my students went through that. Because you build a fundamental understanding of things that somebody just coming at it straight into the DAW doesn’t get.
Going back to Covid, how did it impact your mental model of being in the studio? You must have felt like you could provide the exact tools that musicians needed to accomplish their goals while locked down.
Well, it was very validating, but I spent most of my time dealing with people’s misunderstanding of the laws of physics.
Latency?
Yeah, like I had people say to my face, “No, you’re wrong, I’ve seen it. There were 16 people all around the world all singing the same song at the same time.”
Ah, because of videos on YouTube?
Yeah. There’s a future product in here — I have to be careful about what I say — but there are ways to effectively erase latency in the experience of the people that are in remote locations. If I’m the producer and I’ve got people in Austria and New York and London, I’ve got latencies all over the map. So the basic concept for me is that everybody has to have a solid reference that they can perform to — they cannot listen to each other. We’ve done it before, I know that it works. But there’s some subtleties to making it an actual product that would make it easy like snapping your fingers.
The other thing is that (and JUCE figures into this part) I’m in my studio right now and I want you to do a vocal, and you don’t have a DAW. All you have is an audio interface. But I want you to be able to hear a bit of reverb, so that your performance can be inspired. So that you’re not singing in a dead box.
Yeah, I need my reverb!
That’s another problem we have a way of solving and it’s on the roadmap, but we have to pick our battles.
The latency thing, talkback, the whole thing with Sessionwire’s success is shifting the point of reference. The performer is always broadcasting the reference track to everybody in the session. No latency. They’re recording locally 99.9% of the time on their machine. That’s why we added a remote control. I could control your Logic or ProTools rig while you’re performing and lift that burden off of you.
It sounds like it would be easy to be nerd-sniped into building a remote-first DAW!
No, we have no interest in building a DAW, zero.
That’s probably for the best!
We have our own app on Mac and Windows. That means we have data and audio channels, the ability to control both time domains from a single location.
We’ve always prided ourselves on being “Switzerland.” And people are blown away, “I can use Logic while my buddy is using ProTools?.” I’m like, “Yeah, it’s just audio.”
Like, imagine 3000 mile long XLR cables. That’s what Sessionwire is.
How do you handle support? It sounds like you’re engaging with a lot of people.
Honestly, it’s surprising to me, but our support volume is very low.
There’s sometimes an initial bump where people haven’t quite understood what’s going on. The phenomenon that we see is a lot of very experienced recording engineers overthink it. They think it’s more complex than it actually is.
You have to talk them away from the edge of the cliff and say, “No, no, no, there’s nothing fancy going on here. You don’t need a PhD to use this. You copy that link and send it to your pal. They click on the link and boom — they’re in the studio and they can hear your audio.”
Do you frame that whole flow as a type of UX problem? Or do you guys just do a good job without trying? [laughs]
I don’t know! Our self-reference is, “What do I want?” Knowing what I know about a singer or saxophone player or harmonica player that’s not technically skilled, what is the lowest friction path to get that person into my virtual studio and up and running?
So you are always thinking about that friction. You’re dogfooding, using your own product, you live in your own product. And you’re the boss, so you’re making sure to prioritize that?
Yeah, where does the complexity and the friction of the software end?
Because if you’re in the recording studio — I’ve been in that situation so many times — you have somebody coming in and they’re nervous and they don’t know what’s a “cue mix.” Or, “Oh no, you can’t be that close to the microphone.” You know what I mean? There is a tremendous amount of friction in a recording studio when you have an artist that’s technically inexperienced or inexperienced in that environment.
One of the benefits of Sessionwire is they’re still in their environment. So you’re kind of pulling that out of the equation.
Have you had any conversations with DAW vendors about baking Sessionwire into the DAW? Is that something you dream about? Or is that illegal to talk about? Don’t violate those NDAs…
Yeah, oh I’m going to be careful here. There’s great uncertainty in DAW-land. There’s been a lot of turnover. There’s been some apocalyptic events, with Avid being acquired and so on and so forth. Various different DAW manufacturers have various different levels of NIH (not invented here) syndrome, where they want everything to be proprietary and whatever.
So, there are signs on the horizon that certain DAW manufacturers are becoming more open, but it’s kind of like a sign wave, right? So it fluctuates quite a lot.
The honest truth is that there is a shift going on. This sounds big headed, but I don’t mean it to be: The shift went from being in a recording studio with a console and a two inch machine to the DAW became the center of the universe. With Sessionwire, the DAW is being moved back over to the two-inch machines.
The focus of creativity, the focus of decision-making, the focus of the session proper is in Sessionwire, and the DAW becomes a secondary thing. Oh that’s Logic, Oh that’s ProTools — it doesn’t matter what that is, it’s recording the audio and it’s different for everyone in the session.
This might be a tough question, but what about Gen Z and Gen A? How do you see the new generations of music making evolving? Do you think that the golden age of recording is sun setting or do you think that there’s a new world to still be unlocked?
Yeah, there’s a lot of basement/DIY people. My wife has this app, it’s a karaoke app. And people pay like 30 bucks a month for this thing. She goes, “It’s not because of the karaoke app, it’s because people can make recordings of themselves and hope to be discovered.”
So there’s this whole element, especially in the younger generations, where they’re still thinking that “Oh, I’m so awesome that somebody’s going to discover me and they’re going to pull the freight.”
But there’s still a big chunk of people out there that are like us, willing to try and do what they can to drive their goals and their careers and their visions forward. And this is a great tool for them. It doesn’t matter what the style of music is.
At some point you do this thing and you’re listening back and going “Is this good? I don’t even know anymore.” So wouldn’t it be great to like click on and go? “Hey, what do you think of this?” And they go, “That’s really awesome. But I hate that chord.”
I think there will always be that desire for people to share what they’ve done creatively with other people even if it’s within the personal network.
You were talking about the importance of saying no to features. How did you decide, “Yes, we’re going to build plugins.” Why was that a yes? Was it always part of the plan?
We always were going to build plugins. We always knew what plugins we were going to build. We always knew that we were going to build them with JUCE.
It just came down to a matter of time to market and dealing with internal resource constraints.
The future for us is nuances within the plugins that can massively reduce friction for remote workflows. And that’s automating processes within the DAW environment, such as audio routing — people don’t have to think about that, especially when they’re trying to do something where there’s latency involved. So we have some sort of “Wizard of Oz behind the curtain” stuff that we are going to be doing with the JUCE plugins over the next probably 12 to 18 months.
It’s a no-brainer because, I mean, JUCE takes care of so much stuff that is just time-consuming and horrible. You know, from a development standpoint, you can get right down to getting results and getting feedback from those results. There’s just nothing like it out there.
Robin was interviewed over Sessionwire’s video chat by Sudara from Melatonin.dev.
Have comments on this interview? Ideas for who we should talk to next? Let us know on the JUCE forum.