Hi , I have dedicated 45 years to the professional music industry, operating two major studios specializing in recording classical, modern, and pop music. Throughout my career, I've worked as a composer, producer, and sound technician, and I've enjoyed the success of several major hits across different countries. What we're witnessing now with AI is merely the tip of the iceberg. I'm making a bold prediction: the next evolution in music will be personalized AI music delivered directly through the radio. This innovative AI-powered radio will feature a bot capable of tailoring music to your specific requests-choosing the genre, lyrics, and emotional tone to suit your mood. This custom music will be streamed for you and saved in the cloud, ensuring you can revisit your personalized tracks anytime. The future is clear: digital advancements will revolutionize every aspect of our lives, transforming them entirely."
Your Radio prediction sounds crazy, awesome, and terrifying. I’m not sure if I hope you’re right or wrong. Either way hopefully there will always be a place for live music, and human made music. Thanks for watching!
@@amberelmmusic I have been a fervent lover of producing, composing, and recording pop music, classical music, modern music, and my own pieces since 1975. Despite the rapid changes in the music industry, I am continuously pleasantly surprised. In 1982, I acquired my first advanced drum machine, the LinnDrum, produced by Linn Electronics. This was one of the first drum machines to produce realistic drum sounds. Many musicians thought I was crazy, but it got incredibly busy. Together with a friend, an expert in programming this drum machine, we went from studio to studio. The LinnDrum could sync perfectly with a 24-track Studer or other brands, often resulting in drummers sitting next to us to give directions. Issues such as crosstalk in the drum booth and noise gates on the mixing console became things of the past. A good recording studio cost about $150 per hour at that time. From 1982 onwards, everything changed at breakneck speed, with the introduction of synthesizers with memory and other peripherals that were cheaply released by various companies. Mixing consoles from Studer and Solid State SSL were quickly replaced by new models. Technology is fantastic, but it is unclear when this evolution will stop, just like the continuous changes in the music industry. If I could, I would return to the era of vinyl records. Back then, a musician who sold 50,000 albums could easily earn $200,000. Recently, I spoke with a musician who had 4 million downloads on Spotify but had earned only $4,000. Unfortunately, progress is inevitable. At heart, I am a musician, composer, and producer. Long ago, I was in the Wisseloord Studios with my band, at the same time as Def Leppard. I spent a whole day hearing various drum tracks-just snares and toms-as the musicians strived for the perfect drum sound. That album turned out to be fantastic as well. Image and sound, including post-processing, will forever evolve, and musicians and designers will have to adapt. Just like the drummer who sat behind his drum computer to get the perfect feel, this cycle will continue to repeat. One aspect of music that I firmly believe will endure is the magic of live performances. But can artificial intelligence ever replicate that? In my opinion, no-there is simply no substitute for the energy and connection experienced during a live show.
You are not taking into account the social aspect of music, people like music that other people like too... Already we have infinite music, but people crowd around a few thousabd hits because they bond with it.
There's a scene in Amadeus where Mozart hears a Saltieri piece, then Mozart sits down and makes the Saltieri piece much greater and Saltieri's heart breaks. Gonna be a lot of Saltieri moments coming.
Another thing to remember when talking about how "The big executive won't want to pay money to a human composer ..." ... We are moving towards a world where YOU can become that executive. You can make your own ... music, films etc and you can publish them yourself. So the AI stuff "cuts both ways".
I don't think it's about which one is better, but in in which fields AI music will be preferred and in which it wont. Remember that when recording was invented one would have thought it would kill all musicians. Biut nowadays people can still make a living of performing music, even though technically "it makes no sense" to ever pay for live music (you can listen to that same song on your phone for free) Recording didn't make people stop attending the opera, it made listening to music while commuting something accessible to everyone. And in the process it created other music-related jobs. AI music will replace "functional music". If you are a film student that wants a soundtrack for his short film, or a big company that wants background music for their ad. You probably don't care about the details, you want something "good enough" and as cheap as possible. If the relation 90% of your clients have with the music you make is "it gets the work done". Then yes, probably AI will take your job. And you will have to adapt (maybe live performing, teaching, prompting, etc) AI will replace the Lo-fi hip-hop I put in the background while I work, but it probably won't replace the people playing at my wedding, my favorite content-creator live streaming his songwriting process, buying merch of my favorite bands, etc.
You are spot on ! At least Udio still cam make you feel like a producer ,spitballing ideas ! You can input your own lyrics as well. It's kinda like jamming if you are a song writer ! But yeah,most normal people do not care if AI made it or whoever,as long as it's good ! Anyway,great video mate,Subbed !
Thanks! Ya and the more you try to make something usable, the more you have to prompt it, sort of like trying to get a certain performance out of an artist. The difference is the artist is a freaking computer. Crazy.
ok ... keep it real. AI music will be better than humans can play or make. This is what will happen. We humans make machines that can do stuff better then we can. (its what we do) However ... Humans will always be impressed by seeing other Humans do a skill they cannot. So ... there will always be a market for performing live music to an audience.
For sure. Live music will always have a place. Its just crazy to me what AI generated music can do today based on what it could do just like 5 months ago. It makes me wonder what it will be able to do in a year. And I guess the big bummer is the musicians and composers I think it will put out of a job. Anyways, I could be wrong about all of it…who knows. Thanks for watching!
I think it depends on how it’s presented. If I just posted some of that in a regular video NOT about AI, I think people would just accept it and appreciate it. Knowing it’s AI, makes people listen much more critically and point out everything that a person probably wouldn’t do. Anyways, thanks for chiming in and for watching!
I wouldn't be so sure - Udio actually has copyright filters. However, our studio primarily digitizes analog tape/vinyl material. So Udio does NOT know 99% of ALL music - mostly little and unknown bands. Not 1% of music is available digitally. We have already discovered copyright infringements in music from the 1970s. And not just us.
@@etofok 99% of ALL music is currently only available analogue (unknown bands on vinyl, cassettes). This is therefore never recorded by the AI. We are digitizing this analogue sound carrier - and it would take us another 100 years. There are millions of titles in the publisher's archives that are only available on tape.This remains hidden from the AI forever.
I personally find Udio (and Suno) mind blowing. As a non musician but music aficionado I have wasted hours on the platform. It's almost been therapeutically helpful after long days at work. I can absolutely see having a huge impact for those in the music industry that being said the same will happen to pretty much every job out there... in fact, live music will always (?) remain a thing even in a very AI world so technically an area where there is hope for something. Think about it from a broader perspective. What is currently freakin out artists on many level should be freaking everyone and if anything, as humans, we'll always put some value on human made work even if as a novelty. The same cannot be said about most jobs AI will impact. Also, I have to ask, is that Olafur Arnarlds on your thumbnail for that video ? That's what drew my attention lol
Thanks for watching! I agree with it being mind blowing. I feel like I could make a hundred more videos going into the implications. And yes that is Olafur Arnalds lol. I put him there just because he’s one of my favorite artists, and since he writes for tv on occasion, it played into my point that AI might put his work as a media composer in jeopardy. Thanks for watching!
I'm curious to see how long the gold rush atmosphere lasts. With so many tools coming onto the market for non-musicians (People who can't play an instrument or sing themselves), the platforms are now flooding the market with hundreds of millions of hits. We'll all get rich - assuming everyone buys at least 1,000 titles a day and listens to them. The great sell-off of commercial music has begun. That sounds like a gold rush to many people - but with so much oversupply of all the AI song generators, the price of gold has fallen to 0. The music would no longer be of any value to anyone other than yourself.
I think one big possibility is that services like Udio, are bought up by big media companies. That way they can use the music created by that AI can be used exclusively by them for their various projects. I think that because of the way the ToS reads. But at the end of the day who knows. I’m just hoping we are creative enough to find a way to use it and not be put out of work by it. Thanks for watching.
🤖💔🤖 Years ago no one knew or had the means to edit an image, today we all have an image editing program on our computer, and that does not mean that graphic designers have disappeared. Everyone can edit an image, but only a professional knows exactly how it works. I think the same thing will happen with music.
Yea but you do know there’s a difference between editing an image and generating infinite professional images from a single short text prompt right? This will lower the barrier of entry, and raises the quality expected. So naturally the value of music will go down, just like the value of stock images will go down when you can generate as many images as you want and as it improves get exactly what you. Stock images, stock music tend to be seen as less good than the best artwork and music. Well now you need to be better. There’s still a place for composers but you’re going to need to realize what’s happening to understand where you can fit by knowing what the state of the technology is and how you can ride its wave
acestudio is what comes to mind. midi to audio and audio to midi. i just wish it wasnt subscription based. when i made 9 dollars off of 1000 downloads of one of my songs thru distrokid i gave up money making music anyway at this point. ai might be not helping, but labels and streaming companies have already done musicians in when it comes to profit. composers might be injured but there will be more producers - replacing a person on a track to an ai on a track. but we started doing that with drums 40 years ago, then bass, and its only progressed.
Will the machines listen to AI music when we are no longer here? Is live music going to become the last bastion of the human race...possibly playing music crested by AI? Strange times we are living in. Super and thought provoking video 🤔
Thanks for the comment. 1) live music will defiantly always have a place. 2) maybe the AI will wonder how we can be so creative and will only “want” to listen to human music. Haha. Thanks for watching.
@@amberelmmusic Haha, maybe! Who knows where this will end up, but hopefully there will still be enough people out there who will appreciate the creativity, skill and emotion that a human composer can bring to music. I certainly hope so anyway.
Udio is what got me into making music (and i think i'm putting out some pretty good songs). But it won't do everything for you, you still have to bring something to the table. Still, it has definitely lowered the barrier to creating music.
It’s lowered the barrier for sure. I almost see it like using samples to create a track. Except instead of using others music you are “creating” the samples by promoting it. Thanks for chiming in and for watching!
Udio got you into "making" music? 😂 You are doing shit! I‘m happy for you that you can make a computer generate music for you eventhough your lack of talent but YOU ARE NOT MAKING THE MUSIC! AI DOES IT FOR YOU. Never forget that! You are not a musician by using UDIO! 😂
@@jerrymcpommes8473 -shrug- well whatever I do for 6-10 hours per song is whatever you want to call it. Which results in awesome songs that I own... At the end of the day I have birthed new songs into the world that wouldn't be there if I wasn't here. You can play semantic games as much as you want, doesn't change anything :)
When these music generators update and have more user input control, everyone including professionals will be using it. They will use it or fall behind. People will make great music using these AI generators. There will be music that you never knew was possible. It will be extraordinary to the ear. The problem is AI will probably be better than humans because it will be able to monitor each persons technique and ideas. The AI will be able to combine all the good ideas and make masterpieces. There will be fake human artists that get popular with the AI creating all of their music. The most popular music will be created by AI but people will think that humans are creating it with the help of AI.
It doesn’t even just make music. You can get it to do some of the best text to speech you’ll hear. So expressive! Like audiobooks, speeches, lectures, radio shows and even standup! You can write a script and even add directions to some extent. You can go from taking to singing, change styles in the middle of whatever it is. And so on. ElevenLabs just announced their own so expect more development with the competition starting …. ElevenLabs looks just as good, (some think it sounds better but I don’t see it with their demos) but it appears to be able to give you a whole track with one prompt (rather than 30 second chunks) and judging by the prompts for some of them MAY allow you to prompt for a BPM. I sure hope it can accurately put in a BPM and ideally a key signature . I hope for some more tools for us producers there’s so much more to this than just generating tracks without much musical input. It’s just, like Image AI’s, it’s going to get good at generating good sounding music before it gets really good at giving you more and more microscopically detailed representations of what you ask it to do.
I signed up for eleven labs about a week ago but it wasn’t for music gen. It’s because I was originally going to clone my voice and have the video be narrated by AI (but my voice). Didn’t know it did music generation. I’ll check it out.
Why are people so amazed by ElevenLabs? I’ve heard better stuff in Udio. There’s some cool stuff they have but it’s not THAT much better if you’ve explored UDIO properly, at least going by their announcement demos
Adapt or Die. I adapt. I love using this thing, but I have submitted a couple of pieces I did off site for feedback to working Producers and Artists and one of them stated, "That sounds like Smashing Pumpkins" At that point, I decided not to try to Publish or Claim any Intellectual Property Rights to anything that is created there. For now. Interestingly, (to me) I have found uses for this for Projects other then Original Production and Distribution. Incredibly complex productions and compositions from this Platform. But absolutely no transparency on how it was Trained. Clearly it was Trained with Music that is already available readily and most likely Copyrighted. I think you may be right in terms of people who actually are playing Music for a living. The Writer's Strike did not really win any concessions. I assure you that Scripts are being written with AI. Despite the resolution of the Strike. AI Prompt Engineering does take some experience. Good Analysis.
Thanks for commenting with your perspective. It would be awesome if we could see how these things are being trained, or if there was some mechanism for fairly compensating artists that are being used in the training. That being said, I wonder how different it is from me hearing say a smashing pumpkins song and being inspired by it to write a similar sounding song. As artists we are always gaining our inspirations from other art that exists. Thanks for watching! It’s a topic that is so interesting to me.
@@amberelmmusic Yeah me too. So, in my moderate research of the Legal and Ethical Considerations of the Use of AI; I found that most Companies are putting in place Guide Rails and Frameworks for Training. Among them is Transparency in what Data is used to Train an AI. I think that the EU has already placed this Requirement on AI Apps or Platforms that are Deployed within its purview. The Challenge is that these Platforms and Tools are being Deployed by Companies and Persons outside of the EU Purview.
By the way it's good to know that everything you're making with udio belongs to them. You have no rights at all to make money off it or use it in any commercial setting (podcast games etc) its also stated pretty clear that any future Copyright claims and sues from other musicians will be targeted against you and not udio since you broke the contract. The legal site of this AI shit is a complete mess.
@@ishimurabeats6108 I did not really look at Udio's Terms of Service. Also, I am not giving you any Legal Advise here. I do not know if the Udio Terms of Service prohibit you from using what you Create there in Commercial Productions. Or who even owns any Copyrights to whatever you Create there.
@@ishimurabeats6108This is incorrect. You can’t copyright your music, but there is nothing preventing you from monetizing it, as long as you give credit to Udio. People are already using it for faceless streaming music channels that get monetized by views and advertising.
To me it's a glorified sample generator. no matter how "good" it sounds on first glance, the novelty wears off, and you're left struggling with something that doesn't sound like you at all. i guess it's more interesting than ready-made sample loops though. another thing is that while it never sounds truely great or original, it IS very competent technically and music theory-wise - much more than me in many instances. But if i had to rate the chord progressions and melodies i've generated, i'd say that 50% of it makes sense, and the other 50% is nonsense or failure - with that in mind, all generated songs are mediocre, unless a human remixes/rewrites the output with genuine emotion.
Music has gotten so stale that this might be the thing to jump start the music industry. Get some kid to start playing with this and start creating. Breaking the mold and crusty sludge off of what some people think music "should" sound like. AI technology may be the democratizing factor to give someone the inspiration to try something new. Or it might make things worse and be the end of the world as we know it. One of the two. ehhh
Haha. Ya one of the two. It really does remind me of using samples to make music. It can be lazy, or it can be really creative. I guess lime everything it’s all about how it will be used/abused. Thanks for watching.
They don’t go anywhere. They don’t take you anywhere. In the words of Peggy Lee, “is that all there is? If that’s all there is my friend, then let’s keep dancing!”
I think as it gets better it will start to explore themes, and actually go somewhere. It makes me nervous for things like underscore and license music that a lot of times don’t need to go anywhere. Thanks for watching and chiming in!
