Beyond Generative AI: Building Technology for Musicians

Speaker A

I think this is a reason why there's so few venture backed businesses in music. And because you have all of the difficulties of the same you give all the difficulties that you know for us a a normal a normal ed tech company would face, or consumer tech company would face. And on top of that, you're giving away, you know, some chunk of your your margin to to the industry. So from an investor standpoint, your risk return profile is necessarily worse than you know. Than a standard education product for for instance, or than a standard consumer product. This podcast is produced by Amplitude Ventures, your partner in early stage startups.

Speaker B

Today on Sound Connections, Andrew Carlins, co-founder and CEO of Play Anything, joins us to explore the rise of assistive AI in music, how personalized technology could change the way people learn instruments, and why lowering the barriers to music creation could help bring the next generation of musicians into the market. Andrew, welcome to the studio.

Speaker A

Thank you. I'm so so happy to be here.

Speaker B

Andrew, I'm very excited about this talk today. I I will definitely satisfy my my nerdy side, I'm sure. but for the people who don't understand what that means, let's just set the scene, Andrew. Who are you and what do you do?

Speaker A

And Andrew, I'm co-founder and CEO of Play Anything, we build foundational AI models for musicians and then we wrap them in consumer products. So we have some that give you sheet music, some that teach you piano, you know, that that some that give you MIDI, that genre.

Speaker B

It's interesting. Obviously, you know, AI music generation has been maybe somewhat controversial, everyone would agree, the last few years. so maybe try to explain a tiny bit more detail what you guys specifically do and how you do it, just so we can sort of set the scene in the mux of all these different models that's happening right now.

Speaker A

For sure. So we we actually separate ourselves quite quite drastically from the AI generation space. We we call what we do assistive AI. So to the the step one is someone has to come with some audio that they have, say they are original recording. and many many musicians we find, or many cr creators, composers, they write by playing. and they may or may not be classically trained, they may or may not know how to write notation. And if they do, they tend not really to enjoy the process of actually writing out what they played. and so we we allow these creators to upload their content to to our site and then we give them the sheet music or the guitar tabs or the piano roll or MIDI for their creation, which they could then either learn from, share, or or download for MIDI in in in particular and to and then upload into a DAW and edit thereafter. we also have digital notation software. it's AI native. So for people who don't know all the technicalities of Sibelius or Dorico, they could just type in a prompt. Yep. transpose this up a half step and it does it instantly. And on the piano learning side, we have a mobile app called Anything Piano. And what we realized is a lot of people, a lot of creators wanted like needed a way to visualize and and see what what they had played. Also people come to our platform to learn music. We we have people who come to us who have generated their own songs from us to know and want to learn how to play those. So we have a we have a mobile app that will show you what keys to press on piano and listen to you. And give you feedback on that performance. So that's the overall service provid today.

Speaker B

So since it's a podcast that really focuses a lot on the startups, I'll ask some questions that are more of general nature when it comes to let's say if I were an investor and I would try to figure out what the worth is here. So you're based out of California. you yourself has a very impressive background, including education from Stanford Duke and working at big consulting firms and and raise capital to build this. So so what is the potential of this? Why is this worth pursuing from such a skilled person like yourself and and for the investors backing you?

Speaker A

I think for our founding story is pretty unique. you know, we we actually founded Play Anything. At the time it was called Songscription. Songscription is still our flagship product. but we we we founded the company out of a class at Stanford. So three out of the four of us met in a class. and then we we cold emailed the fourth co-founder because he had written his master's thesis on what has become the the bedrock of our AI models. And it was just fun and we were we were getting along and friends and we all have some different levels of musical background. So we felt like a really cool experience. What we didn't expect is it to take off as quickly as it did. We were lucky that you know, we we got an early investment. Our professor actually believed in us, so gave us some money to begin with and you know, in in in a in a in a school program that's incredibly exciting. And so that's that's why we're all building this and we're we're building it as much for our users as it is for ourselves. which is really cool. and why investors backed us, I I think, I think what they saw, we have a very very strong, even even to date, a very strong technical team. and there's a lot of potential that AI has to change the lives of musicians and music learning for the better. again, I'm separating from the generation side, more on the assistive side. And so lowering the barrier to learn an instrument, lowering the barrier to to get the the curriculum, the pedagogy that you need for your own learning. It could be like taking like simplifying sheet music is actually a really difficult task and historically it's cost hundreds or thousands of dollars depending on the complexity of it. and the the the prospect of giving it to musicians instantly, I think you know piqued the interest of the of our investors.

Speaker B

So what again, I'll start with the big questions, I think it and then go back to the small ones. I think it's just interesting for me to sort of satisfy that curiosity. So what's the end goal of such a company? Is that acquire users at scale? Is that sort of selling the technology? Like what what do you imagine is the way that you will succeed amongst so much, you know, so many companies trying to succeed in this space?

Speaker A

We ambitiously pitch ourselves as the platform for the next billion musicians. And so we're we're we're really focusing while our users spin different different levels of play, from beginners all the way to professionals. what's really interesting for us is because w we are all like late beginner, early intermediate musicians, maybe the co-founders, is building building platforms and products that help that would have helped us when we were first picking up our instruments. and making we when when speaking to people that lot a lot of people in the world want to play an instrument and don't because they perceive it as too hard. or the cost of a teacher is just unaffordable, you know, all these all these really valid reasons. We wanna make the the psychological barrier as low as possible. and then also the economic barrier. and and so, you know, we think that the world is better with more live music. And we hope to build a world where we hope our products go toward toward that end. and that vision think requires investment and hopefully we could grow into our own standalone platform and, you know, exist long term. in terms of how we do that, a lot of our investment to date, we're we're quite early stage, has been in developing state of the art technology. as I mentioned earlier, we we we build our own AI models from from scratch and that's our that's our differentiation. So we're doing things that yeah we're doing things that have never been done before.

Speaker B

Hm. And obviously there's a huge risk to that and and someone must do it. And it's always nice when the company succeed both commercially and technically. So that's you that's a really interesting long journey. but maybe take a a few steps back. Like, as far as I could understand, sort of your master's degree, master's of business administration at Stanford is focused on education, correct?

Speaker A

So interesting. I I did take a lot of education classes and I did enter I I did enter Stanford as a joint degree student with the Master of Education. I I I ended up dropping out of the other degree so I could pursue this so I could pursue play anything full time quicker. but it it wasn't necessarily the case that my MBA was education specific.

Speaker B

No. But but you must have had some sort of curiosity coming into this other than meeting classmates that had a similar music interest and they had an idea. Like what was the lead up to starting this business?

Speaker A

yeah, well I I have I I've I've been really passionate about education, honestly my whole life. I'm pretty sure in kindergarten I I wrote down that I wanted to be a teacher when I grew up. if it wasn't that, it was a dad. and I even even in even when I was in management consulting, I I worked on education projects, not not exclusively. and I've had a non profit for years focused on education, particularly for refugees, low income students and different different US public schools. so education is a real real passion of mine. music is music is in a very literal way how I found my voice. I I grew up with a stutter. I I I'm not sure how much of my background you you you know, but I I grew up with a stutter and when you sing your stutter goes away. And so musical music, specifically musical theater was really important for me to like find my voice and be a confident kid. and so like I I I was music was never my thing. I was always something I enjoyed, but I was never gifted at it. So I always needed the sheet music. And this particular problem space spoke personally to me and realizing the power that music had on a personal level. and then tying it into education was how I ended up in this particular area. My co-founders have slightly different stories. like they were they're they're more on the technical and engineering side, so they were building a product that would help. them in particular. They were trying to learn piano to you know, as like a party trick style. and yeah, put those together and you end up with what song scription was and now playing anything is

Speaker B

It's interesting because it like it's you know, it's not the first obvious pursuit you think of when you think about the age of AI and sort of engineers with incredible sort of technology skills. Like it this it almost seems like a niche in sort of the midst of what Stafford would would be pursuing. So why did you think that that idea in particular was the right one to choose from from all of the possible opportunities right now?

Speaker A

Yeah, look, I'll be I'll I'll be honest with you. We we get this feedback like quite often. and it's it's it's at first it was very frustrating and then I got used to it and I made some jokes. so like the the first feedback we get with a jet you know, pick any random investor is you're building in a really small niche. and while you're doing this, pivot to something bigger and one, it's fun for us, but two, you know, we we put really like When we put the idea out there, like we had a very quick basic MVP. and I recommend this for all founders. Like get your idea out there as soon as possible. we saw we got a couple of dozen hit bytes overnight, like organic traffic. and I think there's real genuine untapped interest in this particular area. I think the the total adjustable market is much, much bigger than people imagine. means the there's there's there's a ton of those cases, like you know, Uber, the taxi market famously in San Francisco is way smaller before Uber than it is now. and that's the founder's belief. It's a little bit radical, that that that there is just this lot of latent demand, but it's what we've been finding from people who upload content to us, a lot of songs like original compositions, et cetera. and so the the I I would push back on the niche point. and and the joke I typically make is look, it might be niche, but as musicians, we definitely know how to scale quite well.

Speaker B

I I got the I got the joke. It was dry, it was nice. Okay. But but it makes sense. And you know, obviously most people would go through family childhood were being exposed to music, music education to some degree, or at least an interest. So I definitely can understand the the reasoning about a potential TAM being being large. Obviously, one of the the biggest hindrances that you mentioned yourself is like the affordability of learning, the accessibility of it. Do do you believe that the the pursuit of learning an instrument will grow significantly in the future? Like what's your thoughts about this?

Speaker A

Yeah, this is I don't think that the number of people learning an instrument is about to explode. but we have seen we've we've seen a couple of things. One, the number of songs composed has exploded as a result of generative AI, for example. and also like like o only in recent years can you like even before generative AI, you you people are composing in their doc. Like That was that was wild thirty, forty, fifty years ago. and so there's an explosion of songs that are composed. You also see an explosion of tastes with platforms like Spotify and Pandora, which is exposing people to niche artists. and only I think somewhere between one to ten percent of all songs are ever, ever written down. Never mind for your particular instrument and your particular level. And so, while I don't think the number of people playing instruments will f you know Is set to explode tomorrow. I I think it'll gradually think it'll you know, I think it'll gradually increase. Hopefully with new interventions and personalized curriculums, or we're building an AI native music tutor. so those those interventions hopefully will increase the the retention of music learners and better cater to their needs and their purposes, make it more affordable as well. however, so so so that is true. And what we've seen is the number of songs that are needing to be transcribed is already skyrocketing. and so for for the purpose of our services, being able to personalize the curriculum and personalize the learning, I I think that is set that we we're we're already seeing an inflection point there. and yeah, yeah, that's that's something we're quite excited by.

Speaker B

So so retention in music education is quite interesting. We recently had a podcast about music education in the space, with the founder Koch Raul, and he talked about they they focus on drummers in particular. And the numbers that he brought is around half the amount of drummers that are playing each year are renewed each year. So like ten million active, ten million new, and so so the the natural churn of a of a learner going and trying to learn and then stopping to learn is is quite high. So Getting getting low churn is is a difficult thing in a subscription model with music education. What's your thoughts on this?

Speaker A

I I think that's absolutely true. and I think that I you know some founders might argue that that drumming is unique 'cause they hit a lot of snares, pun intended there. but I I do think it is I I I think the was it the NAM Foundation? There there's there's some there's some research that that was published a couple of years ago that actually cited the same number. that that the the majority of children who pick up like majority of children will have access to an instrument. through their public school, at least in in the United States. and then a majority of them will will other people will ever pick up an instrument, will drop it before graduating high school. That for that for us is a real tragedy. And when you speak to those adults, to to many adults they who had played, they they express a desire to get back into it. And then they also express this like behavior block of I don't have time I don't think I have the time. finding a teacher is really hard or I'm just not that good. I think with any any education business retention will be will be an issue, our view is Two slash three things, change it. one is personalized curriculum. So our fundamental technology lets you write could let you get any song, even like even a song that you you just composed and you wrote, or the song that your grandfather played and you have a recording of that you would you just really, really, really want to hear it live, and and perform it and make it easy for you. So one is is is personalized curr curriculum. Second is personalized pedagogy. So that's where the AI music tutor comes in. Can we we we we can give you the song you want and we could chunk it up for you. but now let's let's actually personalize the teaching experience and make it hyper focused on you. Also anytime you want, because it's on your phone, at a at a rate that's affordable to you. And by the way, it's not meant to replace teachers. I I I I don't think that that computers will get to the level of having a professional teacher sitting next to you. But I think that a lot of the technical stuff, like are you hitting the right notes? are duration of your notes, I I I think the computer's really good at that stuff. so personalized curriculum, personalized pedagogy, you know, how you like to learn, do you do you like getting feedback quickly? Do you like waiting till the end? All that is AI is poised to do really well at. and and and the third is making it a more social experience. so including gamification, you're able to follow friends and see how see what your friends are up to, learning an instrument, especially as an adult, and as a child to practicing at home is really, really lonely. and you're part of an in school ensemble, which is a lot of fun. But a as an adult, fun practicing piano, no, like having having some kind of social component to it during the six days that your teacher isn't there, or the seven days if you don't have a proper teacher, 'cause you may not be have access to one. for us is a way that we could hopefully increase retention and at the same time make learning music really, really cool to hopefully increase the number of people learning instruments longer term.

Speaker B

Hmm. You you mentioned ten minutes ago about being annoyed when people ask you a question about it being a niche market. so from the outside, when you look at your company's journey, it seems like, wow, that went really fast and they had a lot of momentum. but I'm sure there must be some frustration about this founder journey. Like what are the the biggest challenges that you've met building this company that you maybe weren't expecting or you just didn't couldn't foresee?

Speaker A

I don't even know where to start. There's there's there's a there's there's a lot of like every day there's this new spesser. I think for for for startup founders who are who are watching this, any any st any early stage startup founder that tells you that it's all fun and you know it's it's just sunny is is lying to you. and I'd be happy, please, please reach out over over email or LinkedIn or whatever. I'd be happy to to to share more of a more of an more of an insider view. I feel really passionately that we shouldn't Like as founders, of course we have to market, but we shouldn't over market and we shouldn't shouldn't sell a false narrative. So ton of ton of stress. there's there's like e everything related to getting getting visas. So some of our some of our early employees and one of our founders is not American. And there's a lot of uncertainty that comes with can they enter enter the United States on a you know, on a on a on a proper visa now. And I've never navigated that process as a as the the American co founder CEO. so that that that was that was stressful all the way through deciding on the direction of a company. We had we had a lot of internal debates. do we build this piano learning mobile app? Do we not? If we build it, how should it be shaped? and to, you know, thinking about the future of you know product product product visioning and processes. For example, do we trust our own intuition or do we or do we trust the 10 users that we spoke with? and then how do you navigate when when when our own view or when our intuitive view versus the users that we spoke with don't align. and there's also there's also our risk tolerance and new ethics ethics view. we've taken the perspective of proactively engaging with the music industry. that's something that we know that a ton of music tech startups don't do. it's just the approach that made sense for us. But really at every single corner, there's there's been a lot of tension and I'd be happy to get into any specific that that that you walk. But I I I do want to thoroughly answer your question. It is it is not all fun and g it's it's not all fun in games, although it's it is a lot of fun.

Speaker B

So let's break that down into separate categories. one of the things I'm interested in, I I I don't know too much about capital in San Francisco, but I can imagine when you take on capital as a San Francisco startup or Palo Alto, there's a lot of expectation and r rules associated to that. That must bring its own level of stress, I guess.

Speaker A

So it might actually surprise you. what what at least an another advice for founders is you should really vet to the extent you have the privilege of having multiple offers. you should you should vet your your fut your future investor or your current investors as thoroughly as possible. our investors have very strong reputations and they're also known to be quite quite fair and quite reasonable. And that showed from us even from the beginning of the process. how they engage with us, how they treated us, et cetera. And so given how early stage we are, they give us quite a bit of latitude in how we run things. we also you know, we we have we have monthly updates. There's implied pressure. So we understand where we need to go, but it's not very direct pressure of say, you know, an an investor getting upset if we miss a metric for a particular month. it's more more friendly and you know, just making asking us what support we think we need. So we have a lot of trust from the investors that we know better than than them, especially for our company at our stage.

Speaker B

Hm. So that sounds like you've been able to navigate that sort of cleanly. So so let's get into some of the other things. Like what's been the biggest stressor for you as a founder, like the biggest change that you weren't prepared for?

Speaker A

I so I'm I'm a I come from a non technical background. for for most of the company's life, I was quite literally the only non-technical person on the team. like everyone else was either an engineer or had like over a decade of music notation specific experience and become subject matter expert. And so think like every day for me feels like I'm learning something new. I lead growth. I lead our n legal negotiation side, accounting, investor relations, even even a lot of the HR functions. each each one presented its own its own kind of kind of difficulty and and challenge. And I still don't think I'm the best person at it, but I'm hopefully better today than I was yesterday. That's that's the kind of attitude that I try to adopt and it creates a lot of pressure sometimes. but I'm doing my best. Mm.

Speaker B

What about the music industry? Like, has it been easy to talk with them and and sort of be aligned with them about the purpose that you're building this for?

Speaker A

Absolutely. I think it's because it goes back to what you you pointed out at the beginning, which is that we're different from generative AI. We're what you know, I call assistive AI. and the like you sh you you show this to any musician, you show our products to any musician, they're like, Wow, this is so freaking cool. and many of them, including the industry people that we've spoken with, were like, Wow, I I I wish I had access to this when I was using it and Unlike generative AI, they don't view us as coming into direct conflict with their artists and their artist creation. On the contrary, they've we actually help their artists create stuff faster. We have a ton of professional artists and composers and conductors. Like they're they're they're they're they're using our stuff because it helps them be helps them free up more time for creativity. and so those conversations have been like the the attitude, the approach has been really, really friendly. And they've they've they've embraced the idea of a deal. In terms of the the timeline, I think going into it, I was really like I was really hoping to get deals signed really, really quickly. what I've learned was my expectation was quite naive, takes takes quite some time to get all the approvals necessary, etc. and I you know we recently had a publication come out in Music, Music Ally that we signed your first couple of deals and you know, Spirit, Spirit was mentioned specifically, which were really happy to be working with and I've also learned that the fact that we have some deal signed so so quickly, which was much longer than I expected, is actually like my my my expectation was reverse of reality that the amount of time it's taken is actually really, really l little compared to what typically takes. so that that's another advice I'd I give to to founders is the the industry will take a lot longer than you than than you expect. And Or set up a certain way, yeah, to make sure things yeah, things things work. So just have patience.

Speaker B

What what are the deals that you need to sign in order to operate optimally? Like w what are the things that you need to have in place?

Speaker A

We've been dealing direct deals with music publishers. and those deals both have a component of giving us a right to train on on certain works, and giving us the right to create outputs, based on, you know, derivative outputs. And then we we have some kind of you know, we we pay a certain fee a certain for you know portion of revenue to make sure that artists get paid for their work and both both in terms of creating and in terms of what what what people are asking for. we've also had to do deals with individual artists to get an initial data data training corpus. right, 'cause we've been we've been we've been live for quite some time.

Speaker B

There's some quite an interesting precedent now in the music industry where there's been these talks about train first, ask later, versus the opposite way around, where basically the the sort of the the big incumbents right now are clearly have been training without having permission and then they've been trying to sort of sell that afterwards. And that's of course a angle that takes a lot of capital and there's a lot of risk associated to it. But then the companies that Trying to sort of pursue this clean, especially in the generation side, have basically not been able to provide the amount of quality that needed to in order to be competitive in the market. So obviously your space is is a bit different. Like, what's your thoughts about you know you needing to be patient with your business model? Like, would there be a better business of s hyper scaling it and then asking for permission later? Like, what's your thoughts?

Speaker A

Well I can't answer on the generation side. I think I think the courts I mean, even even for assistive AI, the courts are still out on or or well the jury is still out. I guess I could actually use that literally on on what is what is what is legal versus not. And the juries are are out at different lunches if you're in the United States versus Europe versus yeah, East Asia. I c I so I could only speak from like our particular company's experience. because we are looking in the education space, because our our grand vision is like any song, any level with a live AI music tutor to to to help guide you very personalized. it makes like and and because our goal is to grow to be you know more than a one billion dollar business, so venture, venture scale. we knew from the start we would need to sign contracts in with the with with with the industry. So that led to us proactively approaching them and starting negotiations as soon as soon as possible. I I I I I can't speak about generally. I'm I'm just saying for our particular space, that that that strategy made the most sense.

Speaker B

Hm. But is it possible to build a good business model where you need to sort of pay companies for licensing? do you believe that's that's a model that's actually scalable?

Speaker A

it's a great question. our like I I think this is a reason why there's so few venture backed businesses in music. And because you have all of the difficulties of the same you give all the difficulties that you know for us a a normal a normal ed tech company would face a consumer tech company would face. And on top of that, you're giving away, you know, some chunk of your your margin to to to to the industry. So from an investor standpoint. your risk return profile is necessarily worse than you know than a standard education product for for for instance or than a standard consumer product. that it's just part part and parcel of the space. yeah on on the flip side, because that is the reality, there are are relatively fewer fewer competitors than if we were to go into like as you mentioned earlier less of a perceived niche space with a with a dam that's perceived much bigger. and so I would my my answer would be really unsatisfying. It depends. and for us it makes sense. For us our passion, what we think we're really good at, what we enjoy building for. I c I could tell you a lot of user stories that just make like make me really, really happy and get me out of bed in the morning. that wouldn't be true if we were building in some other space as much. But there there are pros and cons. And one clear con is what what she mentioned. And it's that you will have to like you will have to build a fundamentally better business because at at some scale you will you'll work with the industry. Hopefully the industry helps you. in our case, I'm I'm pretty hopeful that they will. for for example, getting into schools, the first question that many schools ask is, Hey, like, do do we have license clearance for this? And the fact they're able to say yes you do, is it opens up new new markets for us. but in many many spaces it may not be be business helpful and just right becomes becomes a little more difficult to build a venture backable business.

Speaker B

Hm. Do you believe it's a winner takes all type market?

Speaker A

one in particular? Which morph?

Speaker B

So let's say the space for the particular focus that you have, which is sort of notation, education, personalized tutoring, like it's gonna is it gonna be a winner takes all type situation? Obviously that's a simplification of reality for anyone, but but do you think that's the scenario here?

Speaker A

Even though we're based in the US, we have users quite international. And I I'm imagining so while music is a universal language, there are different music systems around the world that have different different notation back backgrounds. And so for example, we are able to notate like traditional or classical Indian music. but I think if we were building a product that was specific to the classical Indian music market, we would like we would We w we'd have far greater penetration there. And so I think I think in the Western in the Western canon, it may not be a winner takes all, it might be two or three. but I think even if there is a concentration in the Western canon, there would still be room for non Western specific platforms to to to leak. Our goal is like To build something for everybody, for every musician. but if I'm looking at the current space in music education, you have simply piano, which focuses on piano but has a guitar offering, and you have Usician, which focuses on guitar, but also right, has other offerings too. I think AI offers the opportunity to make that more of a winners, like a more of a Uber platform. I'm not sure if that will be successful, but I'm I'm I'm benchmarking based on what I think our current user base is and what I see the current status quo in the market is. our our goal, yeah. our goal is to build for everybody. but our mobile app is piano specific. So I guess that that that might give you an indication of you know, we think that you might need different products for different instruments.

Speaker B

Hmm. No, it makes sense. And and it's also the big question is is there gonna be one major sort of provider of these models and is there gonna be several companies that are the sort of the interface? like it so there's a lot of unanswered questions of how this might work. You know, some companies have succeeded, well many companies in the past have succeeded of both having tech and sort of the the the customer product, but oftentimes that's not necessarily the case. So w what is your thoughts about that? Like technology versus penetration in the market and your opportunities and maybe your strengths in in that case. Maybe let's say let's say you double down on enterprise technology and other companies might build UIs or, you know, around your technology product. So you are more focused on the actual technology versus the distribution. Like what where do you position yourself in that equation? So

Speaker A

We to date, we have on our website that we offer an API. to date, we haven't gotten enough demand to really to really offer a offer a really comprehensive one. we've gotten a few inbounds from small companies. and so my my worry on the enterprise API route is that the is that the players in the space existing just the the total the total market there is really hard to justify from a venture venture scale perspective. I think if we were if we had ambitions to be a smaller business, or if we hadn't like if if we weren't venture backed, I think the enterprise route there is there is pretty good room for. and we may pursue that in in the future opportunistically. so the consumer route has been what we like we we thought we were going to be an enterprise company at first and we saw a crazy amount of organic consumer demand. So that let us double down there. And we find that because we have this state of the art technology and we continue to push it, we're able to win different, you know, different niches of consumers and hopefully we're able to generalize to the everyday abric the everyday musician over time.

Speaker B

So so how do you you see the traffic coming in? Like is this an initial interest organic and you needed to push that through digital advertising or like how how do how do these sort of belong together?

Speaker A

Prior to us raising, actually, yeah, prior to us raising our seed round, we were entirely organic. so we got I don't I don't remember the exact number. actually it it must have been over like over a hundred thousand users organically. and this is when we just had songs scription. turns out it it's a really, really big need. and you're right that the first The first mindset, the first thought that many investors have is it's very, very niche. and we we we thought that too, and the market has proven that, you know, we we've been very grateful that it's just just to see that we're we're like we we get users writing us saying, I've been wanting this transcribed for years, or I've had these recordings for for decades and I forgot how to play them. And like these are these are real problems that people face globally. and and that's how we knew that we were solving a real, real pain point. and now because we've gotten an infusion of capital, we're able to invest in different marketing campaigns, Google Ads being one of them, and grow the traffic quicker. But we wouldn't have done that and we internally we have a policy that you you don't scale something that can't survive on its own and you just view we we should pay we should only pay for distribution if we view it as fuel on the fire versus kindling itself.

Speaker B

Hm. Obviously that's always the the big the big discussion any SaaS company has is like how how much do you push when it doesn't sort of survive in its own? Like do we have a very positive ROAS on on on the money we put in? But but when you imagine that, you know, more competitors come and compete in this space and sort of the CAC goes up the of natural reasons, like do do you think there is an opportunity to acquire users at scale in the future in a more saturated space or Is this sort of the golden opportunity right now to take a market position?

Speaker A

really difficult for me to project. I think any company's long term differentiation can't be technical and like you know needs to be community driven and keep community centered. And so we we we we I as I mentioned in the beginning, right, we we view our differentiation today in you know threefold. It's any song, your way of learning and any pedagogy. And then you're not doing it alone. And that third part becomes increasingly important over time. So we have a really big user base. Can we get users in touch with each other? maybe if I'm if I'm if I'm arranging something, if if I'm composing something, I want feedback on it. Today a lot of people go to Reddit for it. I don't like there there's there there's no music musician centered place to give to give feedback. MuseCore has some kind of options, but like we think a platform where musicians can talk and meet with other musicians and Just share their passion could be a really interesting value add to the market and something that's untapped. and on the learning side, similarly, you to you you don't want to feel like you're learning the piano alone. and to build that you need a community of users. And that was what I f what I would envision our longer term differentiation would be if and when the market does become saturated. But our view is we've invested in the technology, we're continuing to invest. And so that that differentiation should give us a good amount of runway.

Speaker B

Hmm. So so community is something incredibly hard to make. Well, you can't really make it. there needs to be some sort of natural cause of it. So often when I hear companies talk about community, it's very hard for me to understand what exactly is the approach they were having or the the effect that it has. to create that on its own. So tr try to sort of break that down for me. H what when you talk about community with you guys, what do you mean and and how does that exist and happen?

Speaker A

So right now, we don't we don't really offer these things yet. we've been focused on building out a really as good of a technology and then as good of a product as we possibly can. soon we will begin building out the community based offerings. the fundamental principle is I I I think on on the learning side, like being able to to or I guess the fundamental need that we're solving is learning is a lot more fun when you're doing it when you feel like you're doing it with other people or you could you could show your progress and so giving giving different ways and there are different ed ed tech examples of how to do that. I I don't want to give a very specific answer because we don't have something live yet. and this this is something that will require a lot of iteration for us to get right. And so I I I I'm leaving it vague because there's there's no specifics we have we have yet. On the on the song inscription side, and a lot of the people we serve there are people who are like hobbyist composers and the process of creating music. you could you could you could go on on Reddit channels and you see people are looking for feedback and thoughts and a way to connect with other people who know about music and music theory. and so just providing a platform where people already exist, or or people people already are storing their musical works, providing a platform for them to connect and give feedback and maybe post their compositions, for example, seems like it could add value. I I would say just like any, any, any CEO who's pitching an idea before it's it's live, I could be like, I could be eating my words in a couple of months. I I hope I'm not. I think the need for us is very clear. and we hear this from some of our users too. And so we're we're we're building toward that app.

Speaker B

Hmm. It's interesting. I and I I love your point about that this needs to be more than technology in order to to really become big. and oftentimes that's sort of the unknown under pressure, I guess, that creates these opportunities. Andrew, it's been such an interesting conversation and it's really interesting to hear about obviously your progress has been quite amazing and and sort of the the the the user base that you've been reporting has also been quite amazing. and you still call yourself an early stage company. So what are the next steps for your company? Like what's what's the growth stages that are you imagining and what does that entail?

Speaker A

Proud to say we've grown over 25% month over month for over 12 months in counting. so this is this is where we hope to be. and the fundamental like what still today, we've been we've been live for just about two years. yeah, in we the various capacities. But by live, I mean we were a class project two years ago. Still the number one reason people leave us is because we don't do what they need us to do, which is transcribe, arrange music for them. so it it it is continuing to push the frontiers of our AI models, making sure, you know, with with these new publisher deals, integrating the data that we're getting from from them and making sure that these models are as good as possible. and I I think that's that's the number one way to unlock growth. The the second is a lot of focus is going on is going I going to to toward retaining users on song scription. so we recently, really, really recently, like it's it's it's really exciting for us, like we just launched an AI native notation editor. And we think that that that will position us you know in in in a similar world to to to like Muse score, the notation editor, Soundslice, NoteFlight, even Dorico Sibelius to some extent, although it's less professionally leaning than than that. and we're excited to like we're we're excited to see people engage, like like bring like text-based editing to the world of music notation and and and see if if people are really excited by it. like it'd be it'd be crazy to think that in five years, instead of having to learn the specific software and dragging, yeah, or instead of an edit taking, you know, an hour to do, it takes five minutes because you're just type typing in a prompt. Like that's really, really cool to build and that's another growth avenue for us. And then the third growth avenue. and this is really the biggest bet we've making as a company now is on the mobile app side. and so we're we're constantly iterating on the mobile app, right? We have new builds going out, I think at this rate once or twice a week, making sure that that platform is something that is really addressing a user need and leveraging our technology to like yeah, to to allow people to learn music better. That's that's that's the fundamental goal. Like that's the reason the whole company wakes up. in in in in the morning. That's that's that's quite ambitious. but it's also quite fun.

Speaker B

Yeah, and it's quite interesting to see and and hear from the sidelines as well. Like this is not necessarily something that I naturally would see as a sort of business opportunity, but as you unfold it, it makes total sense. Like music education and music technology for me is a rather new discussion point. We've had a few episodes about it. But for me it's really interesting to see that okay, there's actually real business here and the incumbents in the past have done really, really great business. That obviously is a big market. but it's a different approach than you would have to a usual tech company, even more different than just a typical music technology c company. So there there's so many things to sort of consider when building this company. Andrew, I'm I'm looking forward to following you for the next couple of years. and and I'm sure even in the next couple of months we'll see a lot of progress and and updates from you guys. so how do you how do we follow you the best? Like what's the best sort of source of information?

Speaker A

well you could you could always go on LinkedIn and you know you could always reach out over email. My email is andrew at songscription.ai. I'd I read all my emails and I'm pretty pretty generous with with meeting with folks because that's that's how I that's how I learned like when I was earlier in my career and yeah so you you should you should feel free to reach out anytime or listeners can as well. and thanks so much for having me on. I I I really enjoyed it.

Speaker B

Thanks Andrew.

Speaker A

Mm-hmm.