Joe from Audible Genius

Originally funded by a Kickstarter in 2012, Joe Hanley not only developed a curriculum to teach synthesizer sound design but also created the cross-platform app Syntorial to deliver it — an innovative blend of video demonstrations, interactive ear training and a bespoke synthesizer.

It's nice to meet you! And thanks to Prashant for connecting us.

Prashant’s the best! We originally met because he’s a Syntorial customer. He reaches out here and there for random reasons, sometimes just to show appreciation. Sometimes he wants to talk about Syntorial at conventions and I'll give him coupon codes.

He's amazing at networking. I first met him at ADC two years ago. He seemed to have had a conversation with every single person attending!

I'm always so jealous of that skill. I do not even have one micron of that skill in me. I'm like the hermit in a cave. I mean, don't get me wrong, I'm way better than I used to be 10 years ago. And in a fixed environment like teaching, I'm good.

Understandable. I like to hide in the background... But you've got your face all over your products!

Well, in the beginning, when I first made Syntorial version one in 2013, I didn't put myself in the videos at all. It was my voice, but you never saw my face. I didn't even say my name at the beginning of the course.

At that time I was like "I'm just going to make this awesome internet product and no one will ever see my face". Now I put myself in there because I recognize the human connection makes it more powerful. People like that.

I did the first few lessons and you're essentially hand-holding me through them. I got used to having you as a companion. It's interestingly parasocial. Very personal. More so than YouTube, I guess because of the participation? It works really well!

Thanks. Before I made it, I was a professional musician, working in studios and stuff. But I also taught piano. It was always one-on-one lessons with adults and kids. I wanted it to feel that way. I wanted it to be real as possible, like I'm a teacher in the room with you.

But you have to do it. It's not just me telling you what to do.

It definitely succeeds at having those one-on-one vibes. I'm building a synth and doing sound design for it, so I'm like "do I even know my stuff? I should go through and test myself!"

Yeah, people who have experience will say "I know a lot of this, but it fills in some gaps."

And more importantly, it's the ear training. They have to recreate a hidden patch as they get deeper in and realize "Oh wow, actually, I wasn't able to do that before, even though I knew how all these controls worked. I wasn't able to make what I was hearing". That's kind of the gap we fill in for people who have experience.

Was the idea born out of ear training? I've taken formal ear training classes, so instantly I was like "Oh! This is ear training for synths."

That was the whole idea. I was working in studios in New York. Since I was the keyboard player, I was always tasked with synths. I could look at a synth and tell you what each parameter did. But I'd be sitting there, in that moment "Hey, Joe, we need a synth bass here." And I could even hear what I wanted. But I'm like, drawing a blank. I had no idea how to bridge what we were hearing to the device, the synth.

But here I am being paid to do this. I talked to different keyboardists and they'd be like, "Oh yeah, that just takes time. Just do it for 10 years and you'll be able to do that." And I'm like, I need to be able to do this. So I set out to make some helper tools. First it was spreadsheets that helped me find presets faster, that didn't work. And then it hit me: I just need ear training.

I'd gone through traditional pitch and rhythm ear training. A ton of it. I was obsessed with ears. So I was a whiz at that. I could figure out the notes and the rhythms I'm hearing in a second. But I realized "Oh, it's the same problem, we just need timbre. Synth-oriented tone training."

It's right up my alley. Back in my university days I was really into Golden Ears. It's like ear training for frequencies.

That was for sure part of the genesis.

You went to Berklee, right? Was that a good experience for you?

It was a great experience. I graduated in '03. Wow 20 years ago. At the time I was there, there was a huge turnover rate after the first year, because you know, you have this idea of like "I'm going to be a musician for my job. Music is fun. I'm going to make this my livelihood!"

But then they realize, like anything, whenever you take something you love and turn it into your job, some joy gets lost inevitably, right? Also, you expose yourself to who the best musicians are. There's some amazingly talented people there and you are like "Oh... nevermind." That first year cleans house and the serious people remain.

When I got there, I thought I wanted to be a jazz pianist. After a year, I realized I do not want to do that. No disrespect for the genre. It's a beautiful form of music. But I was like, this isn't me. And so I ended up spending the last three years doing what they call Professional Music, where you build your own major and just exploring everything I possibly could.

At the end of the day, I think the big differentiator between going to a place like that and learning on your own is you're steeped in this constant music environment, full of people that are really experienced. And you're just swimming in it for years. Overall, it was worth the time.

Did Syntorial exist before the Kickstarter?

In 2011, when I got the idea I just was like: this needs to exist. I want this to exist for me. I was so confident about the idea. I started to make it, but I hadn't made software before. I'd tinkered around with some really basic websites. I dabbled in PHP, but I had very little coding experience. But I'd always been really curious about it.

So I started to learn C++, came across JUCE and just started making it. About a year into it, dedicating my entire free time to this thing, I realized I'm not even done yet. This is going to take me another year to make this. I should probably confirm that people will want to buy this before I dedicate another year of my life to it.

That was the main motivator for Kickstarter. I didn't need much money, I only needed about 5 grand to finish it. And I said, look, if I can get strangers on the internet to give me five grand just for the idea of this, then to me, that means it's worth finishing.

I've got a couple friends who've done Kickstarters — they found success by having some sort of preexisting audience. Did you just go into it cold?

Yeah, it was cold. I started making some YouTube videos on how to recreate some famous synth sounds. I would put that in the forums and mention the Kickstarter campaign. I think it picked up word of mouth just because it was (and it still is) a very novel idea. People love sharing novel ideas.

Plus it didn't need that many backers to get to $5000. We actually raised eight grand. And then, that Kickstarter page just exists on the internet after that. It took a year to finish it. It started to show up in search rankings. Over that year, it built up this newsletter for us so that when we did release it in 2013, it just immediately started selling.

Did you experiment with pricing?

With Syntorial we haven't experimented much with pricing. It's been the same price since the start.

Are you using the royal "we"? Or are there multiple people involved in the project?

It used to be a royal we. Now there's other people, like my dad is one of our developers.

Cool!

Yeah, like six or seven or eight years ago, he sold his company. He's been coding since the 80s. He sold his company and rather than retire, he came and worked as a developer. My wife runs our social media. So yeah, it's become more of an actual we than a royal we.

Last year, you released 2.0 where you completely redid the videos. The UI got a big update. Did the lessons stay the same? Or did they get more streamlined?

Streamlined. The "On Your Own" lessons — at the end of every section of lessons, there's a video lesson where I talk about what you might encounter in other synths, techniques for making certain types of patches, subtle programming techniques — they got way more fleshed out.

What's interesting is the knobs jump between values, right? Because you have to guess these patches exactly. If I made you have to set it just right, that would be tedious. I found I had to tweak a lot of those. Sometimes I was giving you too many values, or maybe I wasn't giving you enough. Tons of tweaks in there.

But the overall curriculum, like the structure of it, what we teach, was mostly the same.

You did this other project on the web, Building Blocks. How do you describe it...beat composition?

Beat composition. That's what I've settled on. People love it. And it works really well. It's just really tough to market the idea. Because we're really combining music theory, ear training and composition in a DAW environment.

If you're making music in a DAW, you usually have to learn all that stuff in different places. And it's rarely in a DAW. Music theory is a great example. You're often learning that on a music staff. But if you are a DAW musician, you almost never use a music staff.

So I really wanted to teach these things to DAW musicians in the environment they actually use to make music.

What are you working on now? Is there some other secret big project coming up or is it mainly nurturing the existing business.

Right now it's nurturing. We're working on a Japanese version of Syntorial 2.0. That one's almost done with the proof reading stage. I'm really excited about that. It's a whole new world of people, a new demographic to explore. Japan's got such a cool synth history.

Give me some juicy opinions on web based development vs. JUCE. What's good, what's bad?

After doing both, I much prefer JUCE. And that's not just because I'm being interviewed by JUCE. The Web Audio API is so much younger, it's so much newer. Every browser's different. There's way less information about how to do it. It doesn't perform as well unless you really know what you are doing and can really milk it.

Whereas with JUCE, I don't have to worry about any of that. I just got to make my product. I knew that the performance would be there. All the backend stuff would be there.

What about from the UI side?

It's CSS, which is like Frankenstein code. I think CSS is so unintuitive. You have to memorize how it works rather than guess how it works. With JUCE it was pretty straightforward. It was all logical. If I didn't know how to do something, nine times out of ten I could just guess and it would work that way.

I was kind of expecting you to say it was easy compared to C++!

I will say, C++ is a little more complicated. There's way more rules and structure of it. Javascript is more freewheeling and I do enjoy that. But there's a reason C++ is the way it is, right? It enforces a certain approach that if you follow makes for more bug-free software. Theoretically.

It's the price you pay. [laughs]

I remember the early days though. The first programming language I learned was C++ and man, that was a tough hill to climb. And it was just so helpful to have the JUCE code itself, being able to peer inside to see how C++ is properly coded.

Let's chat about synth jargon and techniques. Do you think about those things in a meta way? How they evolve over time, or where the terms come from...

You know, synths are an adventure. There's a reason you have this huge modular community. Why else would you choose to spend an enormous amount of money on a thing that fills up a whole portion of your house unless you just love the process of it. It isn't even about the end product, right? It's just about getting lost in a maze of complexity. That's what draws a lot of people to synths.

There's so much to dig into mentally. Such a cool history. But in order to make Syntorial effective, I'm ruthlessly the opposite. I try to think about it only in pragmatic terms with as little jargon as I possibly can. I try to keep it very guided, very specific and pragmatic.

Now, it's a synth, there's got to be jargon. But it can get out of hand quickly. And then it just becomes more about the jargon than it is about the end goal.

When comparing multiple synths, they often differ. Was part of your job boiling down what the essentials are?

In the lessons where I teach pulse width, I mention that some synths don’t have an explicitly labeled “pulse width” parameter. For example, in a wavetable synth, you have to find a wavetable that includes various pulse widths and use the wavetable position parameter to adjust the pulse width.

But I don’t mention that in the first pulse width lesson, because presenting too much additional jargon or scenarios at the onset of a new concept confuses the learning process. That's why those "On Your Own" lessons appear every once and a while. One of the things I do is say "Ok, so you've learned it this way. But here are some other ways that might be presented. Here's some other terms you do genuinely need to be familiar with."

You are formalizing something that is not formal. Curating which things are legitimate, good jargon, and which things are excessive or vary or are unnecessary. Do you think about this role? Musical instruments and genres always seem to go through this process of being formalized, and I feel you are a big part of that story when it comes to synthesizers.

When I first made Syntorial, that was one of the biggest things I had to figure out, right. I had to figure out what do I need to focus on. What do I leave out. At the end of the day, I just want you as a synthesist to do what you want to do, right?

One of the things I did was just look at a ton of synths. What terms are they all using? What parameters? We're talking about jargon, but importantly it applies to which parameters am I even including? What parts am I going to teach, what parts am I going to leave out?

I found that a lot of fun to figure out. And I did feel a big responsibility because I was, like you said, creating this formal structure out of one that doesn't exist yet.

Something like pulse width is so out of left field. It's something that someone randomly found out worked and is now in our bag of tricks. And now you're teaching it!

Synths don't scream "I was made by a musician," right. It was like scientists put it together and were like "Hey, musicians, do something with this."

I used to be someone who wanted a synth that could do anything. But now I'm very much the opposite. I like a limited feature set. I like being forced to get creative within a limited feature set. And it feels better. I prefer to open up a synth that has 200 controls as opposed to 1000.

To me the perfect synth presents itself as relatively simple but secretly has a ton that it can do, right? That's the perfect UI. Knowing what to present, and I don't want to say hide, but what to put behind another tab, put somewhere else. So that you get that easy-ish feeling (as easy as a synth can feel) while being able to make it powerful when you want it to be.

I'm in year 4 of building my additive synth and that's my goal. The reason I was asking about jargon is because you are really helping someone build up their mental model of synthesis, how things are linked and work together. An additive synth isn't like a subtractive — but still, giving someone a lowpass filter with a cutoff knob can make a ton of sense, because it links back to that mental model. So yeah, I live and breathe these compromises...

Yeah, that's tricky. It almost boils down to what do you want the identity of your synth to be? Because it's not like putting in the filter or not putting in the filter is right or wrong, right? It's just like, how do you want people to relate to it? Do you want this to be a new thing? Or do you want some familiarness it in?

I think that was something I was lucky to have when I made Syntorial. A very clear focus on the product identity. Which is not easy to find. But once you have that, questions like that become much easier to answer.

True! I decided I would "derive" the identity. You know, iterate my way there. And I regret every moment! All it means is that I'm full of self doubt!

[laughs] I make it seem like I had this crystal vision from the beginning. And in some ways I did. It was crystal clear. But I had to make a thing to realize it sucked. I went back to the beginning for the curriculum for example, maybe six or seven times.

Same with Building Blocks. I got you know, 20 or 30 lessons in — "Oh dammit, the decision I made back in Lesson 10 has led me to this dead end here. So now I gotta go all the way back to lesson 10 and do that differently and rebuild out from Lesson 10."

Sometimes you can have a really good idea. But then once you make it and use it, something that was not apparent to you before becomes exceedingly apparent.

I just added a feature to Building Blocks that in my mind was going to work great. And then I used it. And I was like, "this is dreadful!" I made this part of the UI and I was having a hard time using it. So I had to scrap a weeks worth of work, start over for that week and redo it.

Well, I'm glad to hear you suffer like the rest of us. [laughs] A week isn't that bad!

[laughs] That's true. Part of me is like "Oh god, I gotta do another week's worth of coding?" And another part of me is like "Ooh, another week's worth of coding!"


Joe was interviewed 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

More "Made With JUCE"

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram