Ep 207: AI Will Allow Us All to Make Music We Enjoy

Resources

Join the discussion: Ask Jordan and Mikey questions on AI and music

Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup

Connect with Jordan Wilson: LinkedIn Profile


The Dawn of AI-Driven Music Generation

With the evolution of music, technological advancements have created a gap between individuals' musical tastes and their ability to create it. Enter Artificially Intelligent (AI) music generation platforms. Technologies are emerging that allow users to generate music from text descriptions or simple cues, fundamentally changing how people interact with music and making the creation process widely accessible.

AI and User-Focused Music Making

Such platforms are not merely aimed at music professionals but are designed to empower the average person. While AI is bound to have a significant impact on the music industry, the focus here is the common user. The technology encourages users to pour their individuality into the tracks they create, using their voice or tapping cues for inspiration. One could argue that the synergy between AI and human creativity is cultivating a whole new way to experience music.

How Does AI Music Work?

Imagine a large language model applied to music creation. Techniques like diffusion and transformers are deployed to generate melodies, lyrics, and vocals, dictated by user input. On user-friendly platforms, no expertise is needed to create polished, professional sounding tracks.

Navigating the Legal Landscape

The AI music generation landscape comes with its challenges and complexities. The rapid evolution of technology means policies and laws around AI and copyright are continually in flux. However, platforms are finding ways to allow paying users to use the music they generate for commercial purposes.

The Future of Music Engagement

AI's role in the music industry is more than a simple trend - it is a significant step towards a shift in how individuals connect with music. Customized music generation, live song creation, and experience with different styles and languages are becoming a reality, driven by generative AI.

Accessible Tools for Musical Freedom

Technologies now allow us to enjoy beats per minute control and fine-grain customization of tunes. They offer different plans to suit different usage patterns, with both free and paid plans. More importantly, they offer a platform for anyone, regardless of musical ability, to express themselves creatively.

Keeping an Open Mind

As AI continues to redefine fields such as music, individuals and businesses are encouraged to approach technology with an open mindset. The possibilities are only beginning to be explored, and the potentials for innovative uses of AI in music creation are immense. It is exciting to imagine what the future holds for this remarkable fusion of technology and art.

At the intersection of AI and creativity lies an opportunity for business growth. Platforms that generate AI music provide an exciting tool to enhance brand identity, advertising, and customer experience. A new era is upon us—imagine giving your business the edge with your own unique, AI-generated soundtrack.

Topics Covered in This Episode

1. Understanding AI-generated music
2. The impact of AI on the music industry
3. Exploring AI potential in music creation
4. Using the Suno platform


Podcast Transcript

Jordan Wilson [00:00:17]:
We talk all the time about different generative AI, you know, text to text or text to photo, but There's something I think in a product that I really personally love that I think we should all listen to, and that's text to music. And, today's, story, today's episode, I think is gonna be a special one. I think you're going to love it. Alright. Welcome. Thank you for joining us, and welcome to Everyday AI. My name is Jordan Wilson, and I am the host. And Everyday AI is for you.

Jordan Wilson [00:00:51]:
It is a daily livestream, podcast, and free daily newsletter helping everyday people like you and me not just learn what's going on in the world of generative AI, but How we can all actually leverage it. Right? That's important because we can hear about all these different tools and techniques and tactics, but It doesn't really mean much if we can't use it to grow our companies and to grow our careers. So that is what everyday AI is all about. And I hope that, you know, whether you're listening on the podcast, this one's gonna be a a a fun one. I guarantee you. But if you're listening on the podcast, make sure as always Check out the show notes. Look in the podcast description. We have links, so you can go join our daily newsletter, other related shows, etcetera.

Jordan Wilson [00:01:31]:
And if you are joining us live on the live stream, Thank you for all all of you, joining us. So, let's get in. And before we talk about how AI can allow us all to make music that we enjoy, Let's first go over the AI news. Alright. So first piece of news actually dropped right after our live show yesterday. We got it in the newsletter, but we gotta talk about it. So NVIDIA has unveiled its chat with RTX. So NVIDIA just launched chat with RTX.

Jordan Wilson [00:02:00]:
It's a local AI chatbot app. The tool utilizes retrieval augmented generation, so RAG, and NVIDIA's TensorRT LLM software. So the chat with RTX helps users analyze. It can help users analyze videos, documents on their local machine, but it does require a PC with an NVIDIA RTX 30 or 40 series GPU. So keep that in mind. And, also, the chatbot leverages or can leverage AI models like Mist enroll or llama 2 to provide near instant responses, making it a promising tool for data analysis and research. So users can integrate different tools and local files, different file formats such as text, PDF, documents, etcetera, and even pull in information from YouTube videos and playlists Into the RTX chatbot for contextual queries. Alright.

Jordan Wilson [00:02:51]:
Next piece of AI news. It's been a busy 24 hours if you follow the space. So, renowned AI researcher Andrei Kaparthy has departed OpenAI for what he said were personal projects. So, Kaparthy is a founding member of OpenAI and has, left, and he announced this on his, you know, Twitter or x page, that he was leaving to focus on personal projects. So, the departure was, amicable As he did announce it himself, and he said there is no issues at OpenAI, and he did praise the team and talked about its exciting road map ahead. So he is widely regarded as one of the most brilliant minds in AI, and he had previously actually kind of second stint at OpenAI. So he worked, Originally at OpenAI, and then he went to Tesla, kind of headed their AI for 5 years, and then just came back to OpenAI. So, there's gonna be a lot of talk around this.

Jordan Wilson [00:03:45]:
We do know that he was working on a project that he was calling Jarvis, presumably with agents or assistants. So I'm guessing I'm I'm, you know, I know this is the new section, but I'm guessing maybe there was, something happening in the, agents or assistant space that, you know, maybe caused them to leave for and Personal reasons. We'll see. Alright. And then last but not least, ChatGPT is getting a more personalized memory. So OpenAI's chatbot application, ChatGPT, is introducing a new feature called persistent memory, Which, this new memory feature will save users time by avoiding the need to repeat information in multiple chat sessions. So Right now, it is being tested in small groups for free and plus users, but will eventually be available to all subscribers. Well, we don't know if it will eventually be, available for free and plus subscribers.

Jordan Wilson [00:04:34]:
Right now, they're at least testing it, with both. So if you, do get this feature, you should receive a pop up, or You can just check-in your ChatGPT settings for a new personalization tab. Also, some users who receive early access are also reporting The memory feature can be toggled on and off in the custom GPT section as well. Alright. So you didn't tune in, Oh, maybe you did. You didn't tune in today to talk about the AI news. You probably tuned in to talk about, how we can use AI To make music we all enjoy. So thank you for joining us live.

Jordan Wilson [00:05:08]:
This one, I think, is gonna be a treat. So to all of our livestream audience, thank you as always for Joining us, Tara from Nashville, Brian from Minnesota. 2nd, Brian from Mississippi Gulfs Gulf Coast. You know, Monica is saying, yes. So much impactful AI news today. Yeah. There's always more, so make sure you go to your everyday AI .com and sign up for that free newsletter. But I'm excited now.

Jordan Wilson [00:05:30]:
We're gonna bring on our guest for today. So please help me. Welcome to the show. There we go. Alright. So we have Mikey Schulman who is the CEO of Suno. Mikey, thank you for joining the show.

Mikey Shulman [00:05:43]:
Great to be here. Thanks for having me.

Jordan Wilson [00:05:45]:
Alright. So tell us tell us a little bit about what Suno is.

Mikey Shulman [00:05:52]:
So we're we're we think of ourselves as a music company, and, I guess, Most of us, listen to music only. So kind of you have Spotify or Apple Music or one of those other services, and you passively stream Music, all day long. Some of the time, you're paying attention. The rest of the time, maybe you're not even paying attention. And there's just so much more to music that We can experience, and so that's what we try to bring to people. You know, I usually point out to people that, the gaming industry is, like, 30 or 50 times bigger Bigger than the music industry, and it's because games are really, really active. And so I think of the company's mission is kind of uncovering the other 49 fiftieths Of all of the ways that we're trying to experience music, creating music yourself is one of them, but there's a lot more that comes after that. So, that's what we do in a nutshell.

Mikey Shulman [00:06:45]:
It's also a lot of fun. If you're a music lover, making music all day and letting people make music all day and Letting people smile with the music they make, it's it's a pretty cool job.

Jordan Wilson [00:06:56]:
So I do I have to get a full disclosure out here. So I've used and I'm not exaggerating when I say this. Obviously, I I re I review a lot of GenAI software. We have people reaching out all the time to get on the show. I say no to most of them. I reached out to Mikey because the 1st time I used Suno, I was like, wait. What? I actually remember running upstairs and showing it to my wife because I'm like, I couldn't believe this. You know, what what has the reaction been, so far, to the product? Because I I I would say it's it's a fairly, unique offering in the new generative AI space.

Mikey Shulman [00:07:32]:
Yeah. I think, you know, by and large, people like it. And, You know, maybe, in some sense, it's surprising, but in another sense, it's not really all that surprising. You know, humans are really wired to want to make music, And somehow as we grow older, we kind of get acculturated to not think to make music. I don't know how many people you know when you ask them, like, do you play an instrument? And they'll say, Oh, you know, I I played a lot of piano growing up. I haven't touched it in in in 10 years, so I don't really make music and, like, I think that's That's really too bad. I think all all you need to do is is, find a 2 year old and and see how that 2 year old thinks about music and and rocking out and And and, just to know how wired we are to do it. So I think by and large, people love it.

Mikey Shulman [00:08:15]:
The other amazing thing is to see people sharing The music that they make, and sometimes it is really whimsical. I'll just give you a a funny anecdote. You know, one of the one of the first things that is stuck in my head of of when we realized how cool this is is, still when it was an internal tool. I went to Starbucks with one of my cofounders, and he puts in, his name is Martin, and his coffee comes out with the name Margu, And we make a song about it. And, like, this moment now is just stuck in our brains, and it's light, and it's funny, and there's levity. And, It's just it's just we get a laugh out of it, and now we relisten to this song. And then there's, like, this totally other end of the spectrum of people just pouring themselves into music, and it's like, Yeah. I don't make music, but I I I I I write well, and I can I can write you a song, and it's gonna come out, and it's gonna be me inside of that song even though I'm not a musician? And so, there's just so much that we are able to bring to people that, that makes them smile and gives them an outlet, and and, Yeah.

Mikey Shulman [00:09:18]:
It's it's amazing.

Jordan Wilson [00:09:20]:
So, Mikey, I'm I'm curious. You know, what led to the creation? Right? Is it with, you know, your own musical background? Is it, you know, maybe you were trying other tools and you really wanted something better? Like, what what led you, you know, and I'm sure you have a very talented team you're working with, but what led you all to say, hey. Let's create something that brings Music to everyone with something as simple as text. What what led to that decision?

Mikey Shulman [00:09:48]:
You know, I, I'll say I'll say a couple things there. I I I am a and Very mediocre musician. And so it's cool to to, build myself tools that make me into a, you know, a slightly better music maker. But, I'll just tell you, like, my my cofounder Martin literally, met his wife at a jam session, and, like, this is this is amazing. Music music brings people together, And there was just this realization that we do not experience all of the ways that we are meant to experience music. And if we can build technologies to, Give people that, then then we can do it. And so, I think the the biggest realization is that music, more than any other kind of media, is has this massive gap between, people's abilities, let's say, the average person's ability to make music versus people's tastes. Most people do not really have the ability to make music.

Mikey Shulman [00:10:40]:
They don't play an instrument. They don't know some complicated production software, but They know exactly what they like. You know, you stop some random person in the street. What kind of music do you like? Oh, I really like rock, but, like, I also really like reggae, and I know those things don't really work and together, like, whatever it is. Right? Everybody's like that. And so I think take text for example. Most people can write, and most people don't really have developed tastes in text. And so, it's that gap between what's in people's heads and what's in people's fingers, let's say, that we are able to fill.

Mikey Shulman [00:11:11]:
But then the last thing I'll say is, I I we try very hard not to think about this as text to music in the sense that we're trying to bring this to everybody, not just professional musicians. And so, I I think the the average person does not think I want to Write some text and out will come music. And I don't know all of the ways in which people wanna interact with this, but this is just like, how can we let the average person make music? And we're we're starting with, You know, text descriptions now, but I I think we have bigger appetites than that in the future.

Jordan Wilson [00:11:43]:
Let's well, let's let's talk about that. You know, what what does the future look like? You know? And whether you wanna, you know, talk us through from from Suno side or just, you know, the industry, like, What does the future of of AI and and music even look like? Right? Because when I, I mean, when I use the product, I personally used it. I've enjoyed it, But I can't even, like, fathom. Maybe because I was so blown away. I can't really fathom, like, what's next either for the product or or or for the industry. Like, where else can this go?

Mikey Shulman [00:12:11]:
Yeah. We we've got a lot of stuff coming. You know? I'll say I'll say a couple things. We wanna let people really pour themselves into the music that they're making. And so Maybe you're a singer, and maybe like me, you're a poor singer, but we can let you kinda sing as inspiration into into the machine and Out will come the the right stuff. Maybe you wanna just, like, tap your fingers on on your desk and and use that as inspiration and cues. So so There's lots of different things more than just, kind of the descriptions and the lyrics that that, that people can do right now. I think of this really as, like, the ability to soundtrack your life.

Mikey Shulman [00:12:48]:
You know? You're just walking down the street. All of the noises you're gonna hear, all of the things that you're gonna see, these are inspiration. And, You know, songwriters today are walking down the street and using that as inspiration, and so why can't I?

Jordan Wilson [00:13:01]:
Do you think that, You know, because one thing I always think about is how different generative AI technologies may eventually impact professional industries. And I think that's been very cut and dry for, you know, as an example, large language models. Right? We've already seen, you know, Unlimited use cases on on how you can use, a large language model to to help grow your company, to help grow your career. What about AI in in the music industry? You you know, I'm I'm I'm sure there's other, you know, softwares and small use cases that have been around for for a long time. But is this something do you see whether it's it's your platform or just, you know, something else? Is this kind of the future of of music production where maybe it's more Generative AI and maybe it's less kind of, you know, human voices or, you know, people making, You know, actual, you know, music on on a board or in in a software program. How how impactful do you think AI will be to this industry?

Mikey Shulman [00:14:05]:
I think AI will be fairly impactful to kinda all parts of the music industry. You already start to see this. This is not new. You know? Things like Cher doing the 1st auto tune on a on a hit record. Like, that's kind of AI. We really don't focus on that. We are focused on making know, the next random person you bump into in in the street, how do we give them more access to music? And so, we are focused on the workflows that That person can use, not that a professional can use. And just an opinion, I don't I don't think we're ever really gonna lose our connection to artists.

Mikey Shulman [00:14:40]:
And And so there's a lot that AI can do to help artists make music faster, get the ideas in their heads into into charting songs. It's just not what we focus on. So I I I know that that will be a big industry. I just think The mission for us is to is to change how the average person is interacting with music, and, I really don't think that the relationship you have with with artists. And I think just just as a as an example, there's a lot of stuff out there of, people doing covers. You know, so it's like, let me let me make a new Taylor Swift song or let me make let Taylor Taylor Swift cover Enter Sandman by Metallica. And While those tend to go viral, they are a little kitschy, and it's because you know Taylor Swift didn't do it. And so you don't have the same connection with her.

Mikey Shulman [00:15:26]:
You know, you have a connection with Taylor Swift because Her music speaks to you for various reasons. And so, Yeah. I just don't think that that is so interesting, to be honest.

Jordan Wilson [00:15:38]:
So, you know, I'm curious, because certain things I can wrap my heads around or or I can wrap my head around. Right? Like, A large language model. Right? A large language model is essentially trained on, you know, the the open Internet, and it's ingested all of this information. So, You know, maybe dipping 1, technical toe in the water here. How does it work for AI music? Because, you know, it not only creates the the actual, You know, I guess the beat or the the the melody, but it also creates the the the words, the lyrics, and the actual voice. How does that actually work? Like, you know, without getting super technical, but how do you train, you know, a model to be able to create great Music, great vocals, and to put it all together.

Mikey Shulman [00:16:22]:
Yeah. It's funny. It actually works very similarly to a large language model. There are kind of 2 main techniques people will use for music. There's diffusion, and there's transformers. We are fans of transformers, Given some of our background doing text, before this adventure, but, I would say the cool thing about audio and music in particular is that it's kind of behind The wonderful worlds of images and text, and so we have the good fortune to kinda look at those domains and figure out what's working and figure out what's not. And, Basically, what we're able to do is focus really hard on how do we represent audio in the right way so that you can Basically discretize it and turn it into, like, little words or little tokens, and we spend all day and all night thinking about that problem and then borrowing from the wonderful open source community All of the great things that have made LLMs work pretty well in text, and, that's been Really good for us. It's made us not have to reinvent the wheel.

Mikey Shulman [00:17:21]:
So it actually does work like an LLM. It's just like a lot of next token prediction. Just these tokens don't necessarily correspond to anything super intuitive.

Jordan Wilson [00:17:31]:
Yeah. No. Yeah. That that that makes sense. Yeah. Thanks for breaking that down. So, Hey. If you're joining us live, we are gonna do a little demo here in a minute.

Jordan Wilson [00:17:38]:
But if you do have a question for Mikey, make sure to get it in. A couple couple quick ones here. So, 2 from Tara. So she's asking, are the audio down marks water or sorry. Are the audio downloads watermarked? And then, also, can the downloads be used for commercial purposes? We'll we'll give you 2 at once.

Mikey Shulman [00:17:55]:
Those are those are great questions. So, in, Taking those in order. We do have the ability to, figure out if a song is ours. You can't hear That, but we definitely do. The legal landscape around AI and copyright is, like, Kinda complex and rapidly changing. So I I'll tell you the way it stands right now, which is that, if you're a paying user, it's $10 a month. You can use Your music for whatever purposes you want. You own it.

Mikey Shulman [00:18:31]:
You can put it on Spotify. If you get a 1000000000 streams and make a lot of money, we don't ask for any of it. You can yeah. I mean, assume assuming you're doing something legal with it, you can do whatever you want.

Jordan Wilson [00:18:42]:
Have have there been people that have maybe launched little, you know, music career so far? Absolutely. Using your platform? Yeah. Okay.

Mikey Shulman [00:18:50]:
Absolutely. And the cool the cool thing is, you know, we're seeing things that It's it's it's we're seeing people put together, like, tiny concept albums. Like, here are 10 little song. Here are 10 1 minute songs That tell a story, and I think that's so cool. I think that's, like, not something that I've really experienced before with music.

Jordan Wilson [00:19:08]:
Alright. Well, let's let's do something cool. Let's let's maybe experience, some some music here, Mikey. So let's, go ahead. So hopefully, everyone can hear this. If you're on the live stream, let us know. We we normally don't share, share live audio, but maybe walk us, a little bit through what we have right here. We have an AI love song.

Jordan Wilson [00:19:28]:
So so tell us tell us about what this is, Mikey.

Mikey Shulman [00:19:31]:
Yeah. So happy Valentine's Day. It's Valentine's Day, and, it's such a good, Opportunity for people to experience more than just listening to music, and so it's a great chance to make a song for someone. So I actually made you a song, Jordan, In honor of today, and and so we can we can go through it together. But, I'll show you this. We have a a little experience where you can just, Enter a few things about your Valentine, and out will pop a lovely song. So I entered a a few things that, I love about, your podcast, your livestream, and about you, and And I made you a song, so hopefully this plays. Let's let's give it a try.

AI [00:20:11]:
Your voice so sweet, it made me wanna fly. Oh, yeah. Your knowledge of ad

Jordan Wilson [00:20:48]:
Oh, man. I love it. I love it. Working in the hairdo too. You know? I know, you know, Mike 4 g might might be jealous of that and One of our other listeners and former guests that I got a comment on my hair. Hey. And and let us know if if you're listening on the live stream. Normally, we don't share audio.

Jordan Wilson [00:21:04]:
Let us know if you heard that. But so That's it it seems so, like, realistic. Right? And that's what I was blown away with, Mikey, the first time I used your platform because I wasn't expecting something that sounded so polished and professional. I I mean, how is this even done?

Mikey Shulman [00:21:24]:
Yeah. It's, you know, when we set out to do this, we said we were gonna do audio the right way, which is kind of, Honestly, building something that looks like GPT, and it was a lot of hard work at the beginning. And at the beginning, the audio did not sound good. You know, we're a little over 6 months after releasing our our 1st model. And if you go back and you listen to those clips, you would not say that it sounds polished or even maybe interesting. So, yeah, it's I I don't know. It's it's been really fulfilling to see some of the progress. I I wonder how you feel about making a song live Yes.

Mikey Shulman [00:21:59]:
Together.

Jordan Wilson [00:22:00]:
Let's do it.

Mikey Shulman [00:22:01]:
Let's do it.

Jordan Wilson [00:22:02]:
We yeah. We planned this. You know? Because I want people to see, you you know, the process because sometimes working with, you know, large language model or generative AI, you know, sometimes you don't always see what happens in the little black box and, sometimes it takes a lot of work. So, yeah, let's let's do it live, Mikey. Let's let's go ahead and make, make some music live.

Mikey Shulman [00:22:21]:
Let's make a Valentine, and we can if there's time, we can make a a song in our regular platform. So it's just for those listening. We have this little kind of form that lets you make a song for your Valentine. You can go to v day song.com, and it and it'll bring you here. So, Jordan, I guess I'm gonna drive for you. Who who who is your Valentine?

Jordan Wilson [00:22:38]:
It's definitely Monica. Okay.

Mikey Shulman [00:22:43]:
And tell me, where did you and Monica meet?

Jordan Wilson [00:22:46]:
So we met in Springfield, Illinois. Alright. And let me I'm gonna I'm gonna, like, Give give everyone. So if you're listening on the podcast, very cool interface each time, you know, Mikey's being prompted with a question And the, you know, the art is really cool, by the way. So, yeah, he's just we're going through this this kind of prompt process. It's very easy. It does it for you. Alright.

Jordan Wilson [00:23:10]:
So we got We got Springfield, Illinois.

Mikey Shulman [00:23:12]:
Any any more color where in Springfield?

Jordan Wilson [00:23:14]:
We could say I mean, we can make it fun. We could say 2nd and 4th. It's kind of kind of like streets, that that we held held true to our hearts, you could say.

Mikey Shulman [00:23:24]:
Amazing. I love that. And what makes Monica so special?

Jordan Wilson [00:23:28]:
Oh, she's she's an amazing woman. I'd say the the number 1, like, supporter of any and everything, not just I do, but anyone in our our friends, families, she's literally the biggest cheerleader that anyone could ask for in their life.

Mikey Shulman [00:23:46]:
Alright. Let's let's see how we do. And this is from Jordan.

Jordan Wilson [00:23:56]:
Alright. This was this was easy. So, literally, Mikey's just doing this live. He's asking me questions, typing in something, something it it seems so easy. Right? It's not even confusing. So now we're getting some, some props From the, the the Suno Valentine machine that it's processing and it's done? My gosh. That was, like, 10 seconds.

Mikey Shulman [00:24:18]:
Let's let's hear it.

Jordan Wilson [00:24:19]:
Let's hear it.

AI [00:24:27]:
Springfield, Illinois, a second foyer. In a crowded place, my eyes locked And I knew right then, you were something Monica, you're an amazing woman. The number one

AI [00:25:07]:
my

Mikey Shulman [00:25:19]:
So intel We'll actually make you 3 if you wanna hear another one.

Jordan Wilson [00:25:22]:
Oh, let's let's hear version 2. My gosh. This is this is fun. This is so fun for me. Alright. We're gonna hear We

Mikey Shulman [00:25:28]:
got one in Spanish. I info.

Jordan Wilson [00:25:29]:
Spanish is good. Let's go. Let's let's do the Spanish one because I think, her family might appreciate that one.

Mikey Shulman [00:25:35]:
Okay. Great. Well, that's embarrassing. Let's go to number 3.

AI [00:25:58]:
To my life for the sunny day In the heart of Springfield where I've found its way. You're up the street. The laughter caught my soul. In that moment,

Jordan Wilson [00:27:11]:
Oh, man. So good. And, like, if you're listening on the podcast alright. I'm, admittedly, I'm a softie. But that especially that first one. You like, you got me in my feels, Mikey. Like, I was like, man, like, it felt even though we just went through a quick, Like, 10 to 22nd prompt, there was something that said, like, you know, in a crowded room, our eyes meet. Right? Like, who knew that generative AI could, like, Make a piece of music so quickly, so personalized, and it it like, I felt that.

Jordan Wilson [00:27:43]:
Right? I'm like, man, that was so good.

Mikey Shulman [00:27:46]:
You know, I think right. You you clearly know Monica really well. You clearly, know what is so special about her. Why shouldn't you be able to do something creative like this, and pour yourself into it even if you don't play an instrument. And, one way to think about generative AI is it gives you the ability to do that.

Jordan Wilson [00:28:04]:
Man, you you know what? Something I've never talked about before is actually on our wedding day. You know, she She she had all these great surprises for me. I tried to do something where, you know, I got some friends and, you know, we kind of sang a song and, you know, I produce something, but I'm like, man, It probably would have been better if I would have had something like this. It it maybe would have sounded better. So alright. We we we have a lot of questions here. So, Great ones too. So Tonya is asking, can you prompt certain styles, when when using the platform?

Mikey Shulman [00:28:36]:
You sure can. I don't know how we're doing on time. I can we can do a song together if you want to.

Jordan Wilson [00:28:40]:
Oh, hey. Let's go. You know?

Mikey Shulman [00:28:42]:
Let's do it.

Jordan Wilson [00:28:43]:
This show, you know, hey. We we we normally cap it on the guest, but we can do another one.

Mikey Shulman [00:28:48]:
Let's, let's just create a song real So, like, I I wanna make a, I don't know, a a funk song about, about AI. How about a happy AI?

Jordan Wilson [00:29:03]:
Cool. And so and so now we're we're outside of the kind of Special Valentine interface. So this is the normal, the normal interface. So you're able to just type in, a simple description, And it's, I mean, it's already done with, like, 2 of them. The speed of this still just boggles my mind. So it generates 4 at At a time.

Mikey Shulman [00:29:23]:
Is that right? It it generates 2. I actually just did it twice, to to to kind of, tease some some new capabilities we have coming out. A new model will have coming out. So I'll just use the new model one. So let's see how how we do.

Jordan Wilson [00:29:34]:
New model.

AI [00:29:40]:
Well, I got a funky circuit running through my code. I'm a digital machine with a boogie in my soul. I'm an ad

AI [00:29:57]:
And maybe I'll just show one one quick thing.

Mikey Shulman [00:29:59]:
I'm gonna hit this remix button, which is gonna let you kinda start to change things in the song. And so let's say I I heard that, and I'm like, it's not exactly what I want. I really wanted, like, more 19 sixties funk.

AI [00:30:09]:
Okay.

Mikey Shulman [00:30:09]:
And and then I'll just and then I'll just I can just redo it and and and Change things, or I could change some of the lyrics that I didn't like or something like that. And the idea is to get people actually involved in the minutiae of making music, But not in the minutiae of how, let's say, a professional musician might do it. So here's the 19 sixties funk version.

AI [00:30:34]:
I got a funky circuit running through my toe. I'm a digital machine with a working in my soul. I'm an air, baby. Yeah. I'm caught up a kind. Got that boob in So

Mikey Shulman [00:30:59]:
So lot lot of fun here.

Jordan Wilson [00:31:01]:
So so good. So good. Alright. We have a couple other questions. I wanna make sure we can get to some of these big ones. So, Rolando is asking what languages does Sunu support?

Mikey Shulman [00:31:13]:
A lot. I can't rattle them off for you, but But probably at least 25, with varying degrees of success. So, English is pretty good. Spanish is pretty good. German is pretty good. And then and then, Hindi is pretty good. Mandarin is okay. And then I think, Japanese is pretty good, and then I'm gonna I'm gonna tail off in my ability to list them all off.

Jordan Wilson [00:31:35]:
Alright. And then, Michael joining us on YouTube. What's up, Michael? So he's asking, can you change beats per minute? So we saw you go in and Quickly change, you know, the genre, and, you know, the decade, but can you change, like, specific down to, like, beats per minute?

Mikey Shulman [00:31:51]:
So, the answer is sort of. You know, right now, we we are working on, like, more fine grain control like that because I think that is a really good way to try to do stuff. But Right now, you can kinda do, like, fast or slow or uptempo, as kind of the descriptive words here. Right now, you can't really say a 142 beats per minute.

Jordan Wilson [00:32:09]:
Yeah. Okay. Alright. So so, Woosie here said he stopped listening for a second and hopped over to Suno to try it. Very cool. So Right now and and correct me if I'm wrong. There you you guys have, like, a pretty good, like, free plan as well. Right? Like, anyone listening can go Sign up, and it's it's it's, tell us a little bit about the free plan.

Jordan Wilson [00:32:28]:
And you said, you know, with the the paid plan, you can, you know, commercialize, you know, if you want to. But can you just tell people, you know, a little bit about the 2 different plans.

Mikey Shulman [00:32:37]:
Totally. The, actually, the first thing I'll say is this Valentine's experience, v day song .com, it's just unpaywalled. So go go go have fun. Go nuts. Our free plan will give you, 5 songs a day. And so, you know, the idea here is to give this to people and really let them experience the joys of making music. GPUs aren't free for us, and so after that, it gets It gets a little bit painful. And so, there's there's 2 paid tiers that give you different access to just how many generations, you can do, but, There's a $10 a month plan.

Mikey Shulman [00:33:10]:
There's a $30 a month plan. Gives you a lot more than 5 a day. It gives you access to all sorts of other things, top up credits, etcetera, etcetera. So, yeah. I mean, people are sometimes actually surprised at at how generous the free the free tier is, and and It's amazing to make people smile, so we're happy to do it. Yeah.

Jordan Wilson [00:33:30]:
I love this. I love this. And, you know, Mike, we could we could do this all day. So if you are listening, Uh-huh. On the podcast, don't worry. We're gonna have all of these URLs and a little bit more about Suno in the show notes, so make sure to check that out. But, Maybe what's what's your one either, you know, piece of advice for people who want to kind of live at this, you know, intersection of AI and music and and bring some creativity to their their lives or maybe just where you think, you you know, the industry is going with with AI and music. Maybe Mikey, what are your kind of parting words for the everyday AI guests?

Mikey Shulman [00:34:10]:
I think yeah. Just thinking about how there are different ways to experience music from how you probably normally do And to keep an open mind because we don't know all of the ways that people wanna do that. And so I think, just going through your normal everyday life and trying to Think that through is is important for the companies like us trying to do it, but also important for everybody who's using it to to find their inspiration. I think yeah. I'm I'm I'm really excited about about the future of AI in music. It's gonna touch so many parts of the industry from

Jordan Wilson [00:34:50]:
Same. And and, you know, I think you mentioned one thing, you know, kind of being the soundtrack to people's lives and and bringing a smile to my face because This one definitely, you know, brought brought a smile to my face, one of my favorite episodes. So, Mikey, thank you so much for joining the Everyday AI Show. We really appreciate your time.

Mikey Shulman [00:35:08]:
So great to be here. Thank you.

Jordan Wilson [00:35:09]:
Alright. And, hey, as a reminder, go to your everyday AI .com. Sign up for the free daily newsletter. We're gonna be recapping and And highlighting all of the things we did, probably we'll be able to, you know, maybe even share share, one of the downloads that we did live here on the show. So Make sure you go sign up for the daily newsletter, and make sure you join us tomorrow and every day for more everyday AI. Thanks, y'all.

Gain Extra Insights With Our Newsletter

Sign up for our newsletter to get more in-depth content on AI