Good video Christian. Actually Logic 11 does work on Intel processors it’s just mainly the stem separation tools and a few of the demanding AI features that are only for M1 and above. Cheers. Andy
I honestly think that _"Not made by A.I.!"_ will soon become a branding tag across many mediums, art forms, and products. A.I. is simply being used as a buzzword to brand a technology/products which directly affects that company's share price value. A lot of things that are branded as A.I. are not really true A.I., they are what was formerly termed "Expert Systems". The main difference being where the dataset for the old Expert Systems was their access to physical data storage, the new Expert Systems now have access to the internet as their database. Even though what A.I. currently produces is pretty good, and it will get better, I still think there will be a sterile element to it. It won't be anything particular that you will be able to discern, or identify, it will just be a feeling that something is off. A little bit like in the 1986 remake of "The Fly" movie. When they teleport the steak through the telepods: it looks just like a steak, smells just like a steak, but - upon tasting - just doesn't taste right. I think A.I. produced material will be very much like this. Of course, in the corporate world, if the executives can get ad, TV, and promotional music/video for free, then they will use it, but it will be noticeable. Just as the likenesses of many famous actors are now being used in film and TV commercials. You recognize them, and they fit the intended part/role, but - as visually close to the real thing they are - they are still easy to identify as being artificially generated. I think professional composers and producers will take a hit initially, especially those not in the top tiers with a canon of work behind them, but once things start to become _"sameyfied",_ I think that brands will again want to stand out from the crowd by having _"Not made by A.I.!"_ materials promoting their products to set themselves apart from their competitors. And then we come full circle. I think with anything new or novel, they go through several phases: excitement, awe, mistrust, rejection, and compromise. I think that A.I. is currently at the mistrust/rejection phase. There will be a backlash, but that backlash will only carry weight if the big players get involved too (top directors, actors, producers, bands and singers, etc.), because let's face it, they too also face being made redundant. Sure, they can sell their "image rights", but most actors/singers do what they do because of the fame and adoration it attracts, and that will dissipate when their fans know it's just a render of that person, not the person themselves. I think there will then be a period of compromise where A.I. gets to do the boring/mundane elements, leaving the human to concentrate on the more creative elements of the production process. The very thing that makes it unique. I could be wrong, but I hope not. P.S. I am not a music composer or producer, I'm just a home hobbyist.
To some extent I agree but I think the thing that will seal it for musicians is them playing live to audiences. Being able to play live is proof of skill.
@@michaelbirtill691 It'll be the same as when people started using drum machines... fake strings, autotune, quantize. At first its lambasted, then it dominates and then it juist settles back into the tool-box of stuff we use.
I'm already on the other side of this. I think AI tools are just tools, and people are hating on it for all the wrong reasons. I'm a 3D animator, and people saying you can't make anything creative with it are being willfully ignorant to the creative process. I'm finding the tools profoundly unique in their ability to create animations with subject matter that could never be made with any other medium, and it takes a lot of time, energy, and technical knowledge to get the best results. I think the tools will become so pervasive that ignorant people who are not familiar, will begin to doubt everything they see, even when someone says it isn't AI. One of the only genuine things that anyone will be confident in, are people who explicitly say they are using AI, but making work that is undeniably unique and artistic. Unfortunately AI has been marketed as a copy machine by technologists who are not artists, because that's the metric they use to identify how well the models have been trained. The reality is, in the hands of an actual artist, you can make exceptionally unique and personal work. I've been repeating this statement for 2 years now: AI lowers the bar for entry, but the ceiling is just as high.
I started producing three years ago, and I cannot agree more with educational value of stem splitter. I have analyzed hundreds of tracka using Demucs, and learned a lot. Welcome addition to Logic.
Gotta love Logic. I've been using it since v1 to make songs and tunes. It's always been 'enough' to allow me to make music, from the days of tapes and Romplers until the incredible stuff built in today. And now, with the latest versions, I made, produced, mixed and released an album -- to zero ears! -- on a laptop from my sofa. Crazy times. It always makes my heart sink when people complain about this stuff. It's incredible. I'd hate to be so jaded and joyless. The bass players may not be of interest, but the new bass instruments most certainly are, they sound great. Even if the built in players play them much better than I can.
THAT, was actually beneficial and inspiring UA-cam video, from someone I believe believes… in such a seemingly negative time in our history, we need more people like Christian speaking truth, with nuance and passion. 🙏🏽
I’m gonna go right out there and say it, stem splitting is the biggest thing to happen to music since sampling. The next step will be accurate splitting of each part of a symphony orchestra. 3rd trumpet for me, as it has the solo in Birdland.
I agree, what people are going to be able to do with this is amazing. ALSO... with no income from streams, or sales, I think this will give people massive freedoms to use the entire canon of recorded music to make something we have never heard before.
Moises software can already split into what Logic does [vocals, drums, bass, other], plus individual drums [kick, snare, hat, toms etc...], keys, strings, acoustic guitar, electric guitar. Having said that, a cursory try of Logic 11 reveals a better algorithm for those basic stems that Moises also does. So I"m thinking Apple has future splitting up their sleeve and don't want to release it until the "other" splitting is more intelligent and sounds better.
You'll be lucky if logic can tell the difference between a trumpet and a saxophone. And how are you going to split three or four trumpets apart? AI has no idea how the Trumpet section was originally arranged. It's a difficult problem for sure. It's much easier to separate an entire drum set as the drums have constant pitches.
@@daveking-sandbox9263 very true perhaps the algorithm or ai will ultimately be smart enough to pick up subtle timbre differences in a trumpet trio imparted by the different players, instruments and mouthpieces perhaps. Though I have to confess when people compare an expensive guitar with its budget version my aging ears can barely perceive a difference.
@@rhythmace1 I thank you for your kind words. However, in retrospect I deeply regret leaving that comment! Because in accordance with the teachings of my Shaolin master, I am prohibited from practicing Pun-Fu outside of my temple! 🇬🇧🙏🏻🇬🇧
Always such Wonderful Brilliant Insightful.... I just think that Apple & no one seems to highlight this... Are not supporting those of us that have dare I say it a nEARLY 5 YEAR OLD iMac and being intel don`t get certain elements that when you buy your Logic Pro X doesn`t say we will not give you certain elements if we (Apple) Make better versions of our computer.... Nothing taking away from your Brilliant & always so well presented content Christian... Alas sometimes... We all need to make a stand sometimes to the wrong !!! Keep on keeping on your great Videos !!!
Thanks for putting together your views Christian. For me, one of the only slithers of hope left is perhaps the public completely turning away from anything that is either genuine AI or branded as AI when it comes to artistic creation. That and proper regulation - i.e making it law to label AI-generated material as such - I believe is one of the only ways forward for creative people that don't want to use AI for generating music.
Great point about AI as a learning tool. About AI, it's technology. Technology is supposed to help us to get the technical stuff out of the way so we can focus on what we want to accomplish. This can indeed mean that a person can turn an idea into music without having to learn how to play an instrument or even how music 'works'. But as long as the expression, the musical story remains authentic I actually don't think that matters. AI can get it technically right, but al AI can really do is reproduce based on statistics. My believe is that our expression is based on more than that. For me, I'm old school so I play instruments and make my own production and engineering descicions. But I'm also excited to hear what people will do with this new technique and how it will broaden our horizon. Last point: when something becomes super easy to do it is no longer interesting. So, if that's true, music will be all about imperfection (= character) and the beautiful mistakes it will bring. Perfect music is boring, even AI can do that :)
Great insights. One thing I have learned that works for me, your mileage may vary, is to make the best of what you have at the time, work for you. I'm still using a mid 2010 Mac Tower with a 12 core CPU and the max amount of RAM. I am still on High Sierra so can't update anything but I make it work. Use what you have and make it work. Wish I could update though as it would be fun to play with new toys.
I've been using Izotope RX on High Sierra for years. It contains the same stem splitter (called "Music Rebalance"), in addition to a good number of other useful tools for cleaning and polishing audio. Give it a try.
Music for me has always been a communication of one or many peoples expression of feeling transmitted through the air. The more you remove the human from the process the worse it gets. Alas,🤔 🎶🎹🎶Play On
They said that about sheet music, they said that about tritones, they said that about recording music, then they said that about electric guitars, then they said that about synths, then they said that about mp3, then they said it about samples, and they said it about digital daws, and here we are, and almost no one in this comment section even owns a handful of real instruments. Ultimately it wont really be true, until everything is just an ai radio station, and at that point people will be so far removed from humanity they wont know the difference unfortunately. Point being they have always said this, and yet we still keep making music.
Indeed... I suspect the connection between sound and emotion has something to do with our abilities to recognise the scream of our child in a melee of other noises. Its not that we hear it per se but that we feel something isn't right. I think it boils down to millions of years of evolution that has left us with nuance hard wired to our feelings, not neccessarily to our full cognitive understanding?
I hadn't even thought of the prospect of using the Stem Splitter for removing music from film/tv clips to dub. I have always wanted to try scoring/writing for picture but never known how to start. I think this is such an exciting tool - Thanks for inspiring as always Christian x
I have been playing with the Stem Splitter - and it's been fascinating. Particularly with drums, being able to hear them in isolation I've noticed so many clever programming tricks or drum hits that I hadn't consciously noticed in the full mix of the track. And I love your idea of taking moving picture and pulling out the music so you can then go in and write your own music with the voices of the actors still in place - hadn't even crossed my mind that would be a use case it for. I mainly used Ableton but I'm definitely going to be pulling out Logic for this!
Thanks, Christian. Just a few words on Intel 3,6 GHz 10-Core Intel Core i9, works fast, but it crashes accidentally too often. For 10 years of using Logic I never experienced this before. Now waiting for update :)
(Alex stands and proudly gives Christain a standing ovation, applauding his description of the continuing need for the human elements not available in AI.) These new AI instruments are quickly going now very loudly reveal who is using real players and who isn't. Beginners and transcribers are going to love the new separation capabilities.
I’m in mid projects and am a bit concerned about bugs in the 11, but tbh I can’t wait to upgrade eventually. I’m on the M1Max and still running in Rosetta.. that’s how nervous I’m about running into problems. So far though most people give it the thumbs up! I subscribed ‘cause great vid, great style and delivery! Thanks for posting.
I was there in the early 80s when MIDI was being implemented: the community screamed that it was going to replace musicians. (Every musician I know uses MIDI) I was there in the 90s when General MIDI was released: the community screamed it was going to replace musicians. And much like when 37% of the population believed camera's were going to take your souls in the 1900s, AI is just another technological advancement that puts fear into so many. AI is here - we have to get over that, and find ways to work with it. Embrace it, and grow with it. It's nowhere near where it is supposed to be or will be. GREAT VIDEO!!!
We need more explicit-filled rants from Christian! I love and can relate to this energy 🤣 Couldn't agree more about the points on mediocrity. I am not sure why you would want your DAW to write your music for you - if you're a musician, composer, producer, ideally you got into it to create your own music in the first place. Some might think of it as a good "jumping off point" but why jump off from mediocrity? To only create more copy-paste mediocrity? The world doesn't need more of that. The world needs to hear from YOU.
Because not everyone is the same person? A singer songwriter isn’t always a pretentious “composer” like the present company. Whose going to pay thousands of dollars for session players just to make a quick pop song demo or a mix tape for a Garage band lol?
@@ghost-user559 I haven't downloaded the new Logic yet myself, but having watched videos, it honestly appears to be a bit of a arrange-for-you-plug-and-play approach. There are already "session" virtual instruments which have existed for years, all the Native Instruments offerings, and many drum kit instruments with pre-arranged midi loops to get you started (for example all the Scarbee kontakt instruments). So this idea isn't new, it's just that it's been consolidated and marketed in such a way. I don't look down on anyone using these tools, but I think if you're opening Logic up to create music, I would assume it's because you want to make music and not have it all arranged for you based on some presets which will invariably sound like anyone else using them. No pretense intended there. All that said, if used to loop and generate ideas over, I think it can be useful. I might actually use it for when practicing instruments, could be helpful to create my own custom backing tracks with changing variables to add into my practice routine and solo over with a bit of randomization introduced to keep me on my toes.
@@jonathanwingmusic I did not mean you specifically, more the composers in general, even Christian despite the theatrical approach is still fairly level headed in the end. However people seem to forget where they themselves were 20 years ago. People who have single pieces of equipment in their studio that is more than multiple years of rent tend to forget that most “musicians” are one month away from homelessness, and this is the norm not the exception. And then you factor in college students or below, who basically might just have a lone instrument to practice in their garage, and suddenly they now have an entire band. For 200$ you can have limitless accompaniment and actionable sheet music for multiple instruments, editable midi, and you have live backing tracks for demos if you are a vocalist or singer songwriter. I understand Christians perspective and yours, but I don’t think either can really emphasize or remember what it is to be “normal” or “beginning”. I would wager most for all of humanity the majority of musicians are “mediocre” statistically. They are usually the ones who teach it, preserve it, and the reason we can even read it.
@@ghost-user559 thanks for the thoughtful response and all great points. I grew up using a 4-track portastudio in the 90s so I never really had access to all these kind of tools we have today. The great thing in many ways is the gap has really been closed, anyone with even a used laptop, $20 midi controller, and a copy of garage band can create music for an entire band or orchestra, something I never would've dreamed possible when I was a kid. So I get that! I think there would be a place for the session instruments for anyone looking to create a demo sketch to show friends or a producer - and even I myself have used loops to sketch out structures so I may try this for that sort of thing, knowing I'd go back and replace everything with my own instruments and playing once I get a song structure built. But to me, ultimately there is value in hearing music that comes from the heart of an individual rather than a computer-generated preset. I think that was my main point - taking the wealth and access point to an extreme, I would rather listen to someone with that used laptop and garage band and no musical training but writes a song from their heart and soul, expressing their unique perspective with authenticity even if they have no clue what they're doing technically, than listen to someone who has an embarassment of riches of studio gear/plugins and extensive musical training but creates music which is soulless - technically precise but devoid of any personality and perspective. Of course either scenario can create mediocre music lol. And I agree that most people of any artistic medium lean toward mediocrity statistically, so that's going to happen no matter what (with or without these tools).
I really love the way you discuss in this video. I am old but very young, and I agree that talented people can always squeeze out something interesting where others just become mediocre. I did not download the base player in Logic11 because I can't imagine that I will ever use it. Anyway, everything is good if someone really enjoys it. To record in a home studio is a wonderful hobby. Logic for everyone! And some people love boring music. I'm a little bit picky myself :) Lots of love from sunny Sweden (this week)
as usual I totally love and am educated by your great videos🙏 thank you so much🎈 if I can mention the S word who would've thought that something even greater could emerge from Spitfire Audio than the essence himself! ...your approach is even more genuine and grassroots than before🤟 cheers
You mentioned you have switched to Cubase from logic. If you haven’t already can you please make a video explaining why and a retrospective on your feelings of both programs?
I often switched between cubase and logic in the 90s and cubase often came out with the newest ideas, but it crashed all the time. On the other hand, logic would take their same ideas, release them, with no more crashes. That's how I ended up staying with Logic!
I’m not sure why so many producers are so anti-AI. It’s like I never knew they were such laggards with no vision. Don’t they realize what this will be like when they give us more control, and when it gets better? It will be what we always wanted out of sample libraries. But apparently they’re going to fight it until they get that, and they’ll they’ll resentfully accept it and use it when they see how good it is.
@@Edbrad It's not about the useful tools for musicians. It's about the ones intended to replace them and/or allow non-musicians to pretend they're musicians. Graphic designers are already losing their jobs and many other trades will follow soon. You think bands are going to pay a session player if AI is an option? What are working musicians and producers supposed to do with their many years of practice when anyone can pay 5 bucks a month to pretend they have the same skills? What are the wider consequences of a few companies stealing the revenue of multiple entire industries? It will also affect art as a whole. Most people won't bother putting 100s of hours into something special when they can just take a shortcut and make what everyone else is making. A.I. was pitched as a tool, but that's not what's actually happening, and we're concerned about those choosing to wield it as a weapon.
Thank you for this! There are different types of "artificial intelligence". I use an AI that dates from the 90s in my less creative work. If an algorithm is able to assist a creator's ability to do things that are the most tedious. I can see AI helping one respond to one's :ear fatigue," for example, with eq on a mix. Session player is alright for a sketch of a song, but otherwise, you're spot on: mediocrity is the enemy. The main thing, though, is that AI has to serve the creator, not be served.
If you’ve got soundtoy’s little alter boy there’s loads of fun you can have with stem splitter and whacking it on the vocal track and either shift it up or down by an octave. My favourite so far is oh! Darling where a really high pitch Paul Mac going mental had me in stitches (yes I am probably quite childish).
Re: stem splitting for education purposes, YES! But...I personally feel I grew so much as a musician actually splitting my hairs while transcribing separate parts out of a recording: it was then headaches that helped me tune my ears and develop my analytical skills. The biggest challenge from AI and technology I think is that comfort and ease is not necessarily conducive to authenticity/imperfect perfection -- getting better as a player, songwriter, producer (vs. perfectly equal to everybody else's 1s and 0s)/happiness. Life is a process, and learning is part of that, and it hurts, and it's beautiful.
The weirdest moment was when playing around with udio and generating orchestral music.. i liked the results. And admitting that was a chilling feeling.
As a Cubase/Nuendo user and someone who overpaid for RX9 I've been using RX9 and SpectraLayers for about 2-3 years now. I am not a composer neither a pro musician, I just do my own stuff. But the amount of knowledge I took out from splitting stems of my favourite bands and songs is staggering. There are songs that have different layers of added stuff that I had never fully heard despite knowing and listening those songs for decades. Even the effects on drums or bass that are just covered by other instruments but add so much to overall sound. This is the way that AI should be used. While I don't mind (I think I just don't really care to be honest) all those AI vocalists, AI song makers and Chat GPT's don't really create anything interesting. It might become interesting when "fixed" by a musician or as an addition to a song, but on its own it's just more of the POP that we are forced down our throats by social media and streaming services. Stem splitting, that's something I am interested in!
Just updated Logic yesterday and separated some famous drum tracks and I agree. We knew these drummers were f'ing awesome but when you hear them ISO'd you just shake your head and realize even more how awesome they are. And I might jam along with it to come up with something new. Fun!
Finally Logic is beginning to look tempting for the 1st time in a long time for reasons other than the price and value... Cubase has come with SpectraLayers built-in for some yrs now though and SpectraLayers has the best stem separation of all of them IMO, it's not even close... I can't count the times that thing has rescued me from poor recordings... That "Unmix" works wonders.
‘Escape competition through authenticity.’ Naval. Brilliant. I believe this is a fundamental truth but I’ve never heard it put quite so succinctly. Worth a google. 👍
Useless update for me personally but glad people are having fun with it. I just want screensets to remember the layout of your track stacks. Would be a fantastic way to zoom around big templates. Oh well.
I have several software products that have stem separation but my favorite for education is RipX Daw. I have been taking lessons on the Bass and I find it to be extremely useful because it also displays the notes that are played. This makes it great for learning new songs and practicing them. Since it can be opened standalone I don’t have to open my Daw to use it. There are a lot of other features that I have not explored.
My favourite thing about AI is that its likely to cause some sort of backlash/rebelion against it. A lot of musicians have slept-walked into AI, using generic lyrics, chord sequences and samples whilst quantising, auto tuning and compressing the life out of every performance for the last 2 decades which has made the transition to AI so much smoother because they were acting like semi automatons anyway. Its like people not caring is a human call centre is replaced with AI because they were only reading out set replies anyway. Hopefully with the idea of young people futures in jeopardy and artist income at risk will force people to fight back with the one thing that AI doesn't have which is emotion and human connection...... until AI learns that as well I think the future of humanity's art is going to go the way of boutique fashion, tradition craftmen and organic food. That it will be the preserve of the upper middle classes as a way to show smugness haha
SpectraLayers Pro 10, when used with Cubase, offers advanced sound manipulation capabilities. Notably, it introduces powerful AI-based features, including unmixing (separating vocals from backing tracks) and improved reverb reduction. Additionally, it provides voice denoising, speech separation, and transcription in multiple languages. The latest version also enhances usability with improved sliders and contextual cut/copy/paste option
I think what you said about authenticity is what will thrive. AI music sure there will be use cases but I think it’ll allow for even more people to have a career in music.
So I’m not the only one that thought the stem separation tool is a great tool for stripping music from scenes of films to practice composing cues. Brilliant!! Crazy minds think alike!!
Cubase has stem splitting. Select an audio region. Click Extensions > SpectraLayers, or go to Audio > Extensions > SpectraLayers. Spectral layers one will only be able to remove vocals but if you get the full version you can separate drums and piano and vocals and guitar and then from there even separate the drums into snare and kick.
Logic 11 works fine on my Intel MacBook, only the stem splitter and your beloved chroma thingy 😉don't appear to work as they need the advanced chip. And, as I prepare to be flamed... I like it. I also have an M1 studio where I have been playing with the AI, and it does what it does, it is a tool to be added to my creative toolset. A tool, not a do-it-all all - keep creating my friends.
Cubase Pro already had a stem splitter built in and it got even better in the latest release (v13), so I you might not need to hop DAWs just for that 🙂 Love the rant.
And then out of the blue Christian pops up Stina Nordenstam, one of my favorit artists! An already great video got even greater! Keep puting out fantastic content!👍🍺
what features you use depends on what you're doing. (it's the eternal mac vs. windows discussion). you choose what fits your needs. I find the session player to be brilliant for saving time and consistency. haven't had time to dig in to chromaglow yet, but what I have seen so far is good. (nothing digital will ever match analog 100%, that's just physics, but can get close), stem splitter can be very good for splitting out base, drums and vocals. but as everything else, it has its limitations.
crazy to see the speed of tools being released in the last year or so. as an editor working on commercials, we pretty much have to incorporate ai for lower level stuff feeling closer to finished on client presentations etc, mainly ai voice over, ai sound clean up pre-mix & extensive ai images for boardomatics. as a solo musician, lots of ai mastering. . whats stopping musicians from generating stuff on udio until they like or can completely re-produce something that came from robots but may be a hit? who knows.
I've long held a strong interest in A.I. Along with Nano Technology. However, as a creative type person, I can't help but feel that whilst there can be benefits, particularly for those who lack music knowledge and creative abilities, it's a negative in the long run. Is it not best to hone one's skills naturally? Then whatever comes out, and what's in your DAW, is you, not a machine.
I will certainly upgrade. I’ve been with Logic since 1994. I am a published artist. I’ve had a top 40 hit etc….i am semi retired and now play in a lounge setting using tracks and I’ve been laboring for two decades recreating the tracks from steely Dan, EWF, Queen etc etc And although it’s a labor of love, ITS LABOR! I have spent weeks on a Queen track and I challenge anyone to find a better karaoke version of my Queen tracks( huge shout out to my bud Daniel Harville for knocking out those Brian May guitars). So here we are ABLE TO RECREATE TRACKS WITH STEM so I can delete the lead vocals and pianos? IM ALL OVER THAT. NOT TO MENTION BREAKING DOWN MY OWN STUDIO RECORDINGS! I have about 20 songs produced and engineered by tge inimitable Mark Endert! Oh Lordy I want this upgrade! (And a groove synthesis 3rd wave to boot!)
AI is a complicated subject and has been around since the 50s. It has moments of excitement followed by a spectacular crashes (this is the 3rd wave). Logic 11 has given us 3 types of AI application. The saturation plug in attempts to replicate classic equipment - I always feel this can be replicated by traditional programming, just needs smart people and the benefit is you can get something better than the original - so a Meh! Stem splitter is a pure tool and I think is where AI is at its strongest and supports the artist. Like image processing it has real benefits and does specific things very well allowing creators to do something original. So thumbs up. The 3rd is the "session" generative AI. Here there are a herd of elephants in the room which just get ignored. To their credit Apple use symbolic (like midi data but probably uses something else) to synthesize audio rather than audio data itself so the results are better and can be manipulated. But you used the perfect word "Mediocre". I trained a model to generate Chinese melodies - it was better than me but only because of be being totally hopeless writing my own! I think generative AI is the crash and burn scenario - It uses massive computing power which is uneconomic (This hardware isn't becoming radically cheaper or faster). Even the cutdown Logic session AI pushes the machine and even the M4 NPU will only improve things slightly. But when running locally there is always a limited resource. In generative AI creativity = temperature (it is a parameter that defines how much or little the most obvious solution will be selected) . Temperature is really randomness - creativity is rarely random. Copyright is always an issue and the newer generative audio tools tend to ignore this (which is why they are free) and some of the recent offerings leave that problem with you. Meta are an exception here as they train with licensed data. Music is largely rules based with considered rule breaking. AI writes its own rules - good luck with that! This was not written by ChatGPT because it contains opinion - is that so bad!! I could go on but...
I have lost a job to AI. I was composing for a metaverse computer game. They have now told me it is gong to be better for marketing if the music is composed by AI and want me to facilitate that for a fraction of the fee rather than compose. They want to ride the hype train rather than have the best results. At the AI music industry moment it is tech bros trying to raise money. They are taking in lots of venture capital money and those investors are going to want to see some kind of profitable business. Is it going to be a viable business model for film makers to subscribe to a service to have AI make their music?
Sorry to hear that man, music is not an appreciated art to the masses. Your idea about a subscription service for music for film sounded outlandish at first, but I think I can see it becoming a reality. I doubt big budget films would do so, but smaller films and Netflix TV show type stuff would definitely do it. At this rate, Media Composing might not be a viable job for much longer.
Great video... I agree entirely. Just glad Apple are not charging us $$'s for Logic 11 as it's mostly 'Band In a Box' meets Toontrack EZ... only from 10 years ago. There are sooooo many features that they could have introduced or improved to make a pro workflow more seamless, but they have decided to make it a desktop version of an iPad toy. It seems like everything they add is mostly a 'light' version 'copy' of someone else's product. In my opinion Logic went off the rails when they fired the E Magic guys and introduced live loops. I started my DAW hobby in the 90's with Cakewalk Pro Audio 6 on a PC. I may well be going back to a PC either with Cubase, Studio 1 or perhaps even the new 'Sonar' (when they finally release it) LOL However, thanks for your ideas on ways to use the stem splitter tool.
On AI: with AI most people can make a decent track. Some people can even make it inspiring, surprising and creative. But what got me into music was not the ability to create a final track, but rather the process where my personal feelings and imagination can be translated IRL into sound. Acoustic instruments provide great interfaces for that thing. Some tech tools can augment the experience, but mostly just interrupts the flow. With AI, I predict people will be more interested in what's "real". I think small gigs where the audience can get close to "real" musicians playing acoustic instruments will sell. The live human voice will sell. There is something in the immediacy in acoustic instruments. The total connection between my musical imagination, my fingers movements and the sounds I hear is something I've never experienced with digital products. And that's a challenge for you as a music tech developer. For example I saw some video excerpts from your string libraries and from what I can tell it sounds amazing with texture and almost a human-like aspect. But, I also see your fingers just sustaining a midi trigger while all this sound magic happens. There is this disconnection between the physical input and what I hear that has so often left me uninspired. Music is a body experience and if you can make AI products that can bring back music from the computer mouse to the human body, I'm first in line.
Maybe not entirely on topic, but I was a big hater on the Logic reverbs as well, especially Chromaverb. Until I saw Beck's Logic Pro Demo Project with his song "Colors" which is riddled with instances of Chromaverb and Space Designer. Worth taking a listen and also seeing how one can mix a song entirely using only Logic's plugins.
New Spectral Layers 10 pro splits stems into 7 - vocal, drums, keyboard, guitar, bass, other, and non unmixed (any leftovers). This is the integration to choose when working with Cubase.
I was relieved in one of your previous videos to hear you were finally moving to Cubase. Cubase has had this "stem splitting" tech for a couple of years now linking to another Steinberg programme called Spectral Layers which allows you to do so much more than just split mixes. Apple has just done the usual catch up and just taken the shiny bit of the tech and pretending they've created a revolution. Whilst trying so hard not to be a Logic hater, Cubase just IS so much better, it always has been. Look forward to see how you get on with Cubase/Nuendo/Spectral Layers.
Chromaverb absolutely is not the worst reverb of all time. But you do, really, really, have to spend a good deal of time with it, noting how the different parameters (particularly Distance) interact with the various reverb models. And also to recognise that some of the models are non-linear and not exactly easy to get on with. But 'Room' for smaller spaces and 'Smooth Space' for larger are good starting points.
Yeah I saw an amazing tutorial just now. It sounds fantastic, but like all things Logic you really do have to dig deeper to get the right setup for each part.
As stem splitting (and all AI stuff) keeps getting more advanced it will be interesting to see if anyone uses it to create something truly interesting or original with it, sort of like how J Dilla was able to create new rhythmic concepts and combinations of samples we hadn’t heard yet. I’m not sure that the ability to remove music from film is such a game changer as I don’t feel like there’s a shortage of visual media for beginners and students to practice on. Nice to have some more options? I’m still waiting to see something that AI does that inspires me, personally. So far, it seems like it’s just providing shortcuts, cutting the need for collaborating with other people out, and lowering the barrier to entry. That’s not all bad, just not that exciting so far. I’m sure we’ll get a tsunami of mediocre music out of it. And maybe, some interesting new music if it’s not overused or if it’s purposefully misused.
I think Spectral Layers One came with Cubase in version 11 and after; and it will do Un-Mix Vocals. You do need the paid version to do more stem versions. Just wanted to share that the Vocal un-mixing is possible with the included version. Works really well for say taking a phone demo of a Vocal and Acoustic Guitar and splitting apart to get control over each element and building your tune from there. As long as it was a decent performance, this eliminates the need to go and record guide tracks. Kinda a nice timesaver if you build songs on the regular.
On harmonics, I tend to add one of two preamp plugins from a great little company called Fuse Audio Labs on most tracks. The VPRE-31A and 37C. Or possibly an Omega Transformer plugin from Kush Audio. Well priced, and they add some mojo as described. But willing to give Chroma-thingy a go. On the AI 'assistants', I think Drummer is brilliant for getting you close really quickly. So I can see me using the bass for that for some quick ideas...then exporting to MIDI fast. Really can't see me using the keys player at all.
Rather similar to when generative fill turned up in Photoshop, the AI things in Logic seem to belong to a different product, for people who don't really want to compose, imagine or just fiddle about. Music is only a semi-conscious activity. Automating it... negates it. The tension is between AI (backward and borrowed) vs human agency will only get starker in creative software ...
AI features are a choice. If you don’t want to use them, then you’re free to make that choice. It’s not inherently good or bad, we apply those labels to it. I’m sure great things will be created using AI tools, music included
I am using 11, I use Bitwig for creation but I am mixing a ton shows I recorded during the semester at the college I work at. seems to be working well. nice little ui changes. haven't tried the keyboardist and bassist those could be fun for writing songy type songs... u know the one
Good video Christian. Actually Logic 11 does work on Intel processors it’s just mainly the stem separation tools and a few of the demanding AI features that are only for M1 and above. Cheers. Andy
Thanks for that correction Andy.
How can I hear your music?
I honestly think that _"Not made by A.I.!"_ will soon become a branding tag across many mediums, art forms, and products. A.I. is simply being used as a buzzword to brand a technology/products which directly affects that company's share price value.
A lot of things that are branded as A.I. are not really true A.I., they are what was formerly termed "Expert Systems". The main difference being where the dataset for the old Expert Systems was their access to physical data storage, the new Expert Systems now have access to the internet as their database.
Even though what A.I. currently produces is pretty good, and it will get better, I still think there will be a sterile element to it. It won't be anything particular that you will be able to discern, or identify, it will just be a feeling that something is off. A little bit like in the 1986 remake of "The Fly" movie. When they teleport the steak through the telepods: it looks just like a steak, smells just like a steak, but - upon tasting - just doesn't taste right. I think A.I. produced material will be very much like this.
Of course, in the corporate world, if the executives can get ad, TV, and promotional music/video for free, then they will use it, but it will be noticeable. Just as the likenesses of many famous actors are now being used in film and TV commercials. You recognize them, and they fit the intended part/role, but - as visually close to the real thing they are - they are still easy to identify as being artificially generated.
I think professional composers and producers will take a hit initially, especially those not in the top tiers with a canon of work behind them, but once things start to become _"sameyfied",_ I think that brands will again want to stand out from the crowd by having _"Not made by A.I.!"_ materials promoting their products to set themselves apart from their competitors.
And then we come full circle.
I think with anything new or novel, they go through several phases: excitement, awe, mistrust, rejection, and compromise.
I think that A.I. is currently at the mistrust/rejection phase. There will be a backlash, but that backlash will only carry weight if the big players get involved too (top directors, actors, producers, bands and singers, etc.), because let's face it, they too also face being made redundant.
Sure, they can sell their "image rights", but most actors/singers do what they do because of the fame and adoration it attracts, and that will dissipate when their fans know it's just a render of that person, not the person themselves.
I think there will then be a period of compromise where A.I. gets to do the boring/mundane elements, leaving the human to concentrate on the more creative elements of the production process. The very thing that makes it unique.
I could be wrong, but I hope not.
P.S. I am not a music composer or producer, I'm just a home hobbyist.
To some extent I agree but I think the thing that will seal it for musicians is them playing live to audiences. Being able to play live is proof of skill.
Isn't that just the same as when Queen used to print this album contains no synthesisers on their albums, only to then use synths on later ones....
@@michaelbirtill691 It'll be the same as when people started using drum machines... fake strings, autotune, quantize. At first its lambasted, then it dominates and then it juist settles back into the tool-box of stuff we use.
@@TheCrowHillCo Indeed. I'm currently writing about AI in education and its exactly the same sort of panic
I'm already on the other side of this. I think AI tools are just tools, and people are hating on it for all the wrong reasons.
I'm a 3D animator, and people saying you can't make anything creative with it are being willfully ignorant to the creative process.
I'm finding the tools profoundly unique in their ability to create animations with subject matter that could never be made with any other medium, and it takes a lot of time, energy, and technical knowledge to get the best results.
I think the tools will become so pervasive that ignorant people who are not familiar, will begin to doubt everything they see, even when someone says it isn't AI.
One of the only genuine things that anyone will be confident in, are people who explicitly say they are using AI, but making work that is undeniably unique and artistic.
Unfortunately AI has been marketed as a copy machine by technologists who are not artists, because that's the metric they use to identify how well the models have been trained. The reality is, in the hands of an actual artist, you can make exceptionally unique and personal work.
I've been repeating this statement for 2 years now: AI lowers the bar for entry, but the ceiling is just as high.
I started producing three years ago, and I cannot agree more with educational value of stem splitter. I have analyzed hundreds of tracka using Demucs, and learned a lot. Welcome addition to Logic.
Gotta love Logic. I've been using it since v1 to make songs and tunes. It's always been 'enough' to allow me to make music, from the days of tapes and Romplers until the incredible stuff built in today. And now, with the latest versions, I made, produced, mixed and released an album -- to zero ears! -- on a laptop from my sofa. Crazy times. It always makes my heart sink when people complain about this stuff. It's incredible. I'd hate to be so jaded and joyless. The bass players may not be of interest, but the new bass instruments most certainly are, they sound great. Even if the built in players play them much better than I can.
Just tried it on an old band rehearsal recording! Wow - it really works well! This is going to be so useful!
THAT, was actually beneficial and inspiring UA-cam video, from someone I believe believes… in such a seemingly negative time in our history, we need more people like Christian speaking truth, with nuance and passion. 🙏🏽
I’m gonna go right out there and say it, stem splitting is the biggest thing to happen to music since sampling. The next step will be accurate splitting of each part of a symphony orchestra. 3rd trumpet for me, as it has the solo in Birdland.
I agree, what people are going to be able to do with this is amazing. ALSO... with no income from streams, or sales, I think this will give people massive freedoms to use the entire canon of recorded music to make something we have never heard before.
Moises software can already split into what Logic does [vocals, drums, bass, other], plus individual drums [kick, snare, hat, toms etc...], keys, strings, acoustic guitar, electric guitar. Having said that, a cursory try of Logic 11 reveals a better algorithm for those basic stems that Moises also does. So I"m thinking Apple has future splitting up their sleeve and don't want to release it until the "other" splitting is more intelligent and sounds better.
You'll be lucky if logic can tell the difference between a trumpet and a saxophone. And how are you going to split three or four trumpets apart? AI has no idea how the Trumpet section was originally arranged. It's a difficult problem for sure. It's much easier to separate an entire drum set as the drums have constant pitches.
@@daveking-sandbox9263 very true perhaps the algorithm or ai will ultimately be smart enough to pick up subtle timbre differences in a trumpet trio imparted by the different players, instruments and mouthpieces perhaps. Though I have to confess when people compare an expensive guitar with its budget version my aging ears can barely perceive a difference.
"One DAW opens... another one closes!"
.."crashes"
I'm genuinely amazed I haven't heard this excellent pun made before.
@@rhythmace1 I thank you for your kind words. However, in retrospect I deeply regret leaving that comment!
Because in accordance with the teachings of my Shaolin master, I am prohibited from practicing Pun-Fu outside of my temple! 🇬🇧🙏🏻🇬🇧
I thought it was "Whenever God closes a DAW, somewhere he opens Windows"
Hold the DAW…Hold the DAWWW
So true. Mediocrity through perfection. You are good. Well spotted and well said.
Exciting yes. Steinberg’s Spectral Layers Pro reveals Apple’s Splitter to be a bit of a toy. And Blackmagic Fairlight has a good one that’s free.
Always such Wonderful Brilliant Insightful.... I just think that Apple & no one seems to highlight this... Are not supporting those of us that have dare I say it a nEARLY 5 YEAR OLD iMac and being intel don`t get certain elements that when you buy your Logic Pro X doesn`t say we will not give you certain elements if we (Apple) Make better versions of our computer.... Nothing taking away from your Brilliant & always so well presented content Christian... Alas sometimes... We all need to make a stand sometimes to the wrong !!! Keep on keeping on your great Videos !!!
Thanks for putting together your views Christian. For me, one of the only slithers of hope left is perhaps the public completely turning away from anything that is either genuine AI or branded as AI when it comes to artistic creation. That and proper regulation - i.e making it law to label AI-generated material as such - I believe is one of the only ways forward for creative people that don't want to use AI for generating music.
Great point about AI as a learning tool. About AI, it's technology. Technology is supposed to help us to get the technical stuff out of the way so we can focus on what we want to accomplish. This can indeed mean that a person can turn an idea into music without having to learn how to play an instrument or even how music 'works'. But as long as the expression, the musical story remains authentic I actually don't think that matters. AI can get it technically right, but al AI can really do is reproduce based on statistics. My believe is that our expression is based on more than that. For me, I'm old school so I play instruments and make my own production and engineering descicions. But I'm also excited to hear what people will do with this new technique and how it will broaden our horizon. Last point: when something becomes super easy to do it is no longer interesting. So, if that's true, music will be all about imperfection (= character) and the beautiful mistakes it will bring. Perfect music is boring, even AI can do that :)
Great insights. One thing I have learned that works for me, your mileage may vary, is to make the best of what you have at the time, work for you. I'm still using a mid 2010 Mac Tower with a 12 core CPU and the max amount of RAM. I am still on High Sierra so can't update anything but I make it work. Use what you have and make it work. Wish I could update though as it would be fun to play with new toys.
I've been using Izotope RX on High Sierra for years. It contains the same stem splitter (called "Music Rebalance"), in addition to a good number of other useful tools for cleaning and polishing audio. Give it a try.
Music for me has always been a communication of one or many peoples expression of feeling transmitted through the air. The more you remove the human from the process the worse it gets.
Alas,🤔 🎶🎹🎶Play On
But that rules out synths.
my laptop just went in the bin lol
Totally agree. Humans are very spiritual and powerful beings. We must always remember this gift and keep it as pure as possible.
They said that about sheet music, they said that about tritones, they said that about recording music, then they said that about electric guitars, then they said that about synths, then they said that about mp3, then they said it about samples, and they said it about digital daws, and here we are, and almost no one in this comment section even owns a handful of real instruments. Ultimately it wont really be true, until everything is just an ai radio station, and at that point people will be so far removed from humanity they wont know the difference unfortunately. Point being they have always said this, and yet we still keep making music.
Indeed... I suspect the connection between sound and emotion has something to do with our abilities to recognise the scream of our child in a melee of other noises. Its not that we hear it per se but that we feel something isn't right. I think it boils down to millions of years of evolution that has left us with nuance hard wired to our feelings, not neccessarily to our full cognitive understanding?
I hadn't even thought of the prospect of using the Stem Splitter for removing music from film/tv clips to dub. I have always wanted to try scoring/writing for picture but never known how to start. I think this is such an exciting tool - Thanks for inspiring as always Christian x
glad somebody has said this about stem splitting. will allow us to learn how to play instruments better understand your favourite drummers swing etc
I have been playing with the Stem Splitter - and it's been fascinating. Particularly with drums, being able to hear them in isolation I've noticed so many clever programming tricks or drum hits that I hadn't consciously noticed in the full mix of the track. And I love your idea of taking moving picture and pulling out the music so you can then go in and write your own music with the voices of the actors still in place - hadn't even crossed my mind that would be a use case it for. I mainly used Ableton but I'm definitely going to be pulling out Logic for this!
You are on absolutely blistering top form, thank you!
Wow, thanks, I actually currently have whooping cough so it was quite an editing job!
Thanks, Christian. Just a few words on Intel 3,6 GHz 10-Core Intel Core i9, works fast, but it crashes accidentally too often. For 10 years of using Logic I never experienced this before. Now waiting for update :)
I needed my Christian fix. And this was a good one.
(Alex stands and proudly gives Christain a standing ovation, applauding his description of the continuing need for the human elements not available in AI.)
These new AI instruments are quickly going now very loudly reveal who is using real players and who isn't.
Beginners and transcribers are going to love the new separation capabilities.
I will defo be using the stem splitter, love the way you have presented this new version 11
I’m in mid projects and am a bit concerned about bugs in the 11, but tbh I can’t wait to upgrade eventually. I’m on the M1Max and still running in Rosetta.. that’s how nervous I’m about running into problems. So far though most people give it the thumbs up! I subscribed ‘cause great vid, great style and delivery! Thanks for posting.
I was there in the early 80s when MIDI was being implemented: the community screamed that it was going to replace musicians. (Every musician I know uses MIDI) I was there in the 90s when General MIDI was released: the community screamed it was going to replace musicians. And much like when 37% of the population believed camera's were going to take your souls in the 1900s, AI is just another technological advancement that puts fear into so many. AI is here - we have to get over that, and find ways to work with it. Embrace it, and grow with it. It's nowhere near where it is supposed to be or will be. GREAT VIDEO!!!
We need more explicit-filled rants from Christian! I love and can relate to this energy 🤣 Couldn't agree more about the points on mediocrity. I am not sure why you would want your DAW to write your music for you - if you're a musician, composer, producer, ideally you got into it to create your own music in the first place. Some might think of it as a good "jumping off point" but why jump off from mediocrity? To only create more copy-paste mediocrity? The world doesn't need more of that. The world needs to hear from YOU.
Because not everyone is the same person? A singer songwriter isn’t always a pretentious “composer” like the present company. Whose going to pay thousands of dollars for session players just to make a quick pop song demo or a mix tape for a Garage band lol?
@@ghost-user559 I haven't downloaded the new Logic yet myself, but having watched videos, it honestly appears to be a bit of a arrange-for-you-plug-and-play approach. There are already "session" virtual instruments which have existed for years, all the Native Instruments offerings, and many drum kit instruments with pre-arranged midi loops to get you started (for example all the Scarbee kontakt instruments). So this idea isn't new, it's just that it's been consolidated and marketed in such a way. I don't look down on anyone using these tools, but I think if you're opening Logic up to create music, I would assume it's because you want to make music and not have it all arranged for you based on some presets which will invariably sound like anyone else using them. No pretense intended there.
All that said, if used to loop and generate ideas over, I think it can be useful. I might actually use it for when practicing instruments, could be helpful to create my own custom backing tracks with changing variables to add into my practice routine and solo over with a bit of randomization introduced to keep me on my toes.
@@jonathanwingmusic I did not mean you specifically, more the composers in general, even Christian despite the theatrical approach is still fairly level headed in the end. However people seem to forget where they themselves were 20 years ago. People who have single pieces of equipment in their studio that is more than multiple years of rent tend to forget that most “musicians” are one month away from homelessness, and this is the norm not the exception. And then you factor in college students or below, who basically might just have a lone instrument to practice in their garage, and suddenly they now have an entire band. For 200$ you can have limitless accompaniment and actionable sheet music for multiple instruments, editable midi, and you have live backing tracks for demos if you are a vocalist or singer songwriter. I understand Christians perspective and yours, but I don’t think either can really emphasize or remember what it is to be “normal” or “beginning”. I would wager most for all of humanity the majority of musicians are “mediocre” statistically. They are usually the ones who teach it, preserve it, and the reason we can even read it.
Well said my dude
@@ghost-user559 thanks for the thoughtful response and all great points. I grew up using a 4-track portastudio in the 90s so I never really had access to all these kind of tools we have today. The great thing in many ways is the gap has really been closed, anyone with even a used laptop, $20 midi controller, and a copy of garage band can create music for an entire band or orchestra, something I never would've dreamed possible when I was a kid. So I get that! I think there would be a place for the session instruments for anyone looking to create a demo sketch to show friends or a producer - and even I myself have used loops to sketch out structures so I may try this for that sort of thing, knowing I'd go back and replace everything with my own instruments and playing once I get a song structure built.
But to me, ultimately there is value in hearing music that comes from the heart of an individual rather than a computer-generated preset. I think that was my main point - taking the wealth and access point to an extreme, I would rather listen to someone with that used laptop and garage band and no musical training but writes a song from their heart and soul, expressing their unique perspective with authenticity even if they have no clue what they're doing technically, than listen to someone who has an embarassment of riches of studio gear/plugins and extensive musical training but creates music which is soulless - technically precise but devoid of any personality and perspective.
Of course either scenario can create mediocre music lol. And I agree that most people of any artistic medium lean toward mediocrity statistically, so that's going to happen no matter what (with or without these tools).
I really love the way you discuss in this video. I am old but very young, and I agree that talented people can always squeeze out something interesting where others just become mediocre. I did not download the base player in Logic11 because I can't imagine that I will ever use it. Anyway, everything is good if someone really enjoys it. To record in a home studio is a wonderful hobby. Logic for everyone! And some people love boring music. I'm a little bit picky myself :) Lots of love from sunny Sweden (this week)
as usual I totally love and am educated by your great videos🙏 thank you so much🎈
if I can mention the S word who would've thought that something even greater could emerge from Spitfire Audio than the essence himself! ...your approach is even more genuine and grassroots than before🤟
cheers
“The potential to educate here is outstanding” - absolutely agree. The “Kiss” example was fascinating. Thanks mate 🎸
You really sum up my own feelings on AI, especially regarding the creative arts. Thank, Christian!
Well done! Thank you for your insightful and honest review.
You mentioned you have switched to Cubase from logic. If you haven’t already can you please make a video explaining why and a retrospective on your feelings of both programs?
Yes, please.
I often switched between cubase and logic in the 90s and cubase often came out with the newest ideas, but it crashed all the time. On the other hand, logic would take their same ideas, release them, with no more crashes. That's how I ended up staying with Logic!
Using AI to separate the voice from the background "noise" in the singer's microphone, was a fabulous solution for mixing a live recording.
I’m not sure why so many producers are so anti-AI. It’s like I never knew they were such laggards with no vision. Don’t they realize what this will be like when they give us more control, and when it gets better? It will be what we always wanted out of sample libraries. But apparently they’re going to fight it until they get that, and they’ll they’ll resentfully accept it and use it when they see how good it is.
@@Edbrad It's not about the useful tools for musicians. It's about the ones intended to replace them and/or allow non-musicians to pretend they're musicians. Graphic designers are already losing their jobs and many other trades will follow soon. You think bands are going to pay a session player if AI is an option? What are working musicians and producers supposed to do with their many years of practice when anyone can pay 5 bucks a month to pretend they have the same skills? What are the wider consequences of a few companies stealing the revenue of multiple entire industries? It will also affect art as a whole. Most people won't bother putting 100s of hours into something special when they can just take a shortcut and make what everyone else is making.
A.I. was pitched as a tool, but that's not what's actually happening, and we're concerned about those choosing to wield it as a weapon.
Removing backgrounds in photoshop/figma is very similar. Takes a lot of faff out of something that is so so useful.
Thank you for this! There are different types of "artificial intelligence". I use an AI that dates from the 90s in my less creative work. If an algorithm is able to assist a creator's ability to do things that are the most tedious. I can see AI helping one respond to one's :ear fatigue," for example, with eq on a mix. Session player is alright for a sketch of a song, but otherwise, you're spot on: mediocrity is the enemy. The main thing, though, is that AI has to serve the creator, not be served.
If you’ve got soundtoy’s little alter boy there’s loads of fun you can have with stem splitter and whacking it on the vocal track and either shift it up or down by an octave. My favourite so far is oh! Darling where a really high pitch Paul Mac going mental had me in stitches (yes I am probably quite childish).
hahaha brilliant use! can't wait to try this thanks for the mad genius idea
Re: stem splitting for education purposes, YES! But...I personally feel I grew so much as a musician actually splitting my hairs while transcribing separate parts out of a recording: it was then headaches that helped me tune my ears and develop my analytical skills. The biggest challenge from AI and technology I think is that comfort and ease is not necessarily conducive to authenticity/imperfect perfection -- getting better as a player, songwriter, producer (vs. perfectly equal to everybody else's 1s and 0s)/happiness. Life is a process, and learning is part of that, and it hurts, and it's beautiful.
Really nice insights Christian! You are a terribly smart chap.
(Apart from taking this long to abandon logic…)
You are at a very high musical level. Will follow you for sure .
The 'No...' at 5'40" is the best response to the whole so-called innovation... and sums it all up!
so new musicians learning from these new tools is a bad idea?
They all have their uses.
Spot on about the role of technology in music. New things should inspire us not scare us.
The weirdest moment was when playing around with udio and generating orchestral music.. i liked the results. And admitting that was a chilling feeling.
The only aspects of Logic Pro 11 that require Apple silicone are the AI functions. I have Logic Pro 11 installed on my 2017 iMac.
I think this is one of the best , most insightful videos I have seen in a long time....Im subscribing now!
As a Cubase/Nuendo user and someone who overpaid for RX9 I've been using RX9 and SpectraLayers for about 2-3 years now. I am not a composer neither a pro musician, I just do my own stuff.
But the amount of knowledge I took out from splitting stems of my favourite bands and songs is staggering. There are songs that have different layers of added stuff that I had never fully heard despite knowing and listening those songs for decades. Even the effects on drums or bass that are just covered by other instruments but add so much to overall sound.
This is the way that AI should be used.
While I don't mind (I think I just don't really care to be honest) all those AI vocalists, AI song makers and Chat GPT's don't really create anything interesting. It might become interesting when "fixed" by a musician or as an addition to a song, but on its own it's just more of the POP that we are forced down our throats by social media and streaming services.
Stem splitting, that's something I am interested in!
Just updated Logic yesterday and separated some famous drum tracks and I agree. We knew these drummers were f'ing awesome but when you hear them ISO'd you just shake your head and realize even more how awesome they are. And I might jam along with it to come up with something new. Fun!
Love the passion...Love the delivery...Love the perspective and love the imagination...💯%👈
Finally Logic is beginning to look tempting for the 1st time in a long time for reasons other than the price and value...
Cubase has come with SpectraLayers built-in for some yrs now though and SpectraLayers has the best stem separation of all of them IMO, it's not even close... I can't count the times that thing has rescued me from poor recordings... That "Unmix" works wonders.
I use Artificial Insemination in my compositions almost daily, couldn't live without it
Goooooooshhhh I love your brutal honesty 🙌🏼
‘Escape competition through authenticity.’ Naval.
Brilliant. I believe this is a fundamental truth but I’ve never heard it put quite so succinctly. Worth a google. 👍
Love this bloke. Says it all and half of it is beeped out. Great.
Ya, the drum iso on the Kiss is rad. Great idea on splitting stems for the demo usage.
Useless update for me personally but glad people are having fun with it. I just want screensets to remember the layout of your track stacks. Would be a fantastic way to zoom around big templates. Oh well.
I saw that coming a mile away, glad i switched to studio one i don't need AI bloatware in my daw
Ludite.
AWESOME REVIEW CHRISTIAN!!! LOVE IT 👊
I have several software products that have stem separation but my favorite for education is RipX Daw. I have been taking lessons on the Bass and I find it to be extremely useful because it also displays the notes that are played. This makes it great for learning new songs and practicing them. Since it can be opened standalone I don’t have to open my Daw to use it. There are a lot of other features that I have not explored.
Great video stem splitter has huge creative possibilities couldn’t care less about session musicians, you cant beat the real thing!
great point ! Thanks for your enlighting review!
My favourite thing about AI is that its likely to cause some sort of backlash/rebelion against it. A lot of musicians have slept-walked into AI, using generic lyrics, chord sequences and samples whilst quantising, auto tuning and compressing the life out of every performance for the last 2 decades which has made the transition to AI so much smoother because they were acting like semi automatons anyway. Its like people not caring is a human call centre is replaced with AI because they were only reading out set replies anyway.
Hopefully with the idea of young people futures in jeopardy and artist income at risk will force people to fight back with the one thing that AI doesn't have which is emotion and human connection...... until AI learns that as well
I think the future of humanity's art is going to go the way of boutique fashion, tradition craftmen and organic food. That it will be the preserve of the upper middle classes as a way to show smugness haha
SpectraLayers Pro 10, when used with Cubase, offers advanced sound manipulation capabilities. Notably, it introduces powerful AI-based features, including unmixing (separating vocals from backing tracks) and improved reverb reduction. Additionally, it provides voice denoising, speech separation, and transcription in multiple languages. The latest version also enhances usability with improved sliders and contextual cut/copy/paste option
Thank you, exactly the Logic Pro 11 review I needed!
I think what you said about authenticity is what will thrive. AI music sure there will be use cases but I think it’ll allow for even more people to have a career in music.
So I’m not the only one that thought the stem separation tool is a great tool for stripping music from scenes of films to practice composing cues. Brilliant!! Crazy minds think alike!!
Cubase has stem splitting. Select an audio region.
Click Extensions > SpectraLayers, or go to Audio > Extensions > SpectraLayers. Spectral layers one will only be able to remove vocals but if you get the full version you can separate drums and piano and vocals and guitar and then from there even separate the drums into snare and kick.
best review ever!! respect
I never would have thought to use stem splitter to extract speech from picture, to then rescore. That’s going to be great for practice
Logic 11 works fine on my Intel MacBook, only the stem splitter and your beloved chroma thingy 😉don't appear to work as they need the advanced chip. And, as I prepare to be flamed... I like it. I also have an M1 studio where I have been playing with the AI, and it does what it does, it is a tool to be added to my creative toolset. A tool, not a do-it-all all - keep creating my friends.
Wow! Stina Nordenstam! Great choice.
I just loved that intro! I agree completely- logic and their team don’t care about anything anymore. I blame the tim apple team culture 😂
Cubase Pro already had a stem splitter built in and it got even better in the latest release (v13), so I you might not need to hop DAWs just for that 🙂
Love the rant.
Absolutely true
Yup! Just checking we’re talking about spectral layers right ?? But yes Cubase has had this feature ages ago
And then out of the blue Christian pops up Stina Nordenstam, one of my favorit artists! An already great video got even greater! Keep puting out fantastic content!👍🍺
what features you use depends on what you're doing. (it's the eternal mac vs. windows discussion). you choose what fits your needs. I find the session player to be brilliant for saving time and consistency. haven't had time to dig in to chromaglow yet, but what I have seen so far is good. (nothing digital will ever match analog 100%, that's just physics, but can get close), stem splitter can be very good for splitting out base, drums and vocals. but as everything else, it has its limitations.
crazy to see the speed of tools being released in the last year or so. as an editor working on commercials, we pretty much have to incorporate ai for lower level stuff feeling closer to finished on client presentations etc, mainly ai voice over, ai sound clean up pre-mix & extensive ai images for boardomatics. as a solo musician, lots of ai mastering.
.
whats stopping musicians from generating stuff on udio until they like or can completely re-produce something that came from robots but may be a hit? who knows.
Christian… you’re on fire here! Absolutely immense video. (Let’s get those mashups and alternate film trailers out there…)
I've long held a strong interest in A.I. Along with Nano Technology.
However, as a creative type person, I can't help but feel that whilst there can be benefits, particularly for those who lack music knowledge and creative abilities, it's a negative in the long run.
Is it not best to hone one's skills naturally? Then whatever comes out, and what's in your DAW, is you, not a machine.
I keep wondering as I watch this video - what the hit is on your electricity bill to keep that big synth turned on!
I will certainly upgrade. I’ve been with Logic since 1994. I am a published artist. I’ve had a top 40 hit etc….i am semi retired and now play in a lounge setting using tracks and I’ve been laboring for two decades recreating the tracks from steely Dan, EWF, Queen etc etc And although it’s a labor of love, ITS LABOR! I have spent weeks on a Queen track and I challenge anyone to find a better karaoke version of my Queen tracks( huge shout out to my bud Daniel Harville for knocking out those Brian May guitars).
So here we are ABLE TO RECREATE TRACKS WITH STEM so I can delete the lead vocals and pianos? IM ALL OVER THAT. NOT TO MENTION BREAKING DOWN MY OWN STUDIO RECORDINGS! I have about 20 songs produced and engineered by tge inimitable Mark Endert! Oh Lordy I want this upgrade!
(And a groove synthesis 3rd wave to boot!)
More excellent comment, thank you Christian 👍
this was great...balanced views with outrage... all the feels... :)
that new instrument at the end sure does sound dope! waiting for this one from Crowhill!
AI is a complicated subject and has been around since the 50s. It has moments of excitement followed by a spectacular crashes (this is the 3rd wave). Logic 11 has given us 3 types of AI application. The saturation plug in attempts to replicate classic equipment - I always feel this can be replicated by traditional programming, just needs smart people and the benefit is you can get something better than the original - so a Meh! Stem splitter is a pure tool and I think is where AI is at its strongest and supports the artist. Like image processing it has real benefits and does specific things very well allowing creators to do something original. So thumbs up. The 3rd is the "session" generative AI. Here there are a herd of elephants in the room which just get ignored. To their credit Apple use symbolic (like midi data but probably uses something else) to synthesize audio rather than audio data itself so the results are better and can be manipulated. But you used the perfect word "Mediocre". I trained a model to generate Chinese melodies - it was better than me but only because of be being totally hopeless writing my own! I think generative AI is the crash and burn scenario - It uses massive computing power which is uneconomic (This hardware isn't becoming radically cheaper or faster). Even the cutdown Logic session AI pushes the machine and even the M4 NPU will only improve things slightly. But when running locally there is always a limited resource. In generative AI creativity = temperature (it is a parameter that defines how much or little the most obvious solution will be selected) . Temperature is really randomness - creativity is rarely random. Copyright is always an issue and the newer generative audio tools tend to ignore this (which is why they are free) and some of the recent offerings leave that problem with you. Meta are an exception here as they train with licensed data. Music is largely rules based with considered rule breaking. AI writes its own rules - good luck with that! This was not written by ChatGPT because it contains opinion - is that so bad!! I could go on but...
I have lost a job to AI. I was composing for a metaverse computer game. They have now told me it is gong to be better for marketing if the music is composed by AI and want me to facilitate that for a fraction of the fee rather than compose. They want to ride the hype train rather than have the best results.
At the AI music industry moment it is tech bros trying to raise money. They are taking in lots of venture capital money and those investors are going to want to see some kind of profitable business. Is it going to be a viable business model for film makers to subscribe to a service to have AI make their music?
Sorry to hear that man, music is not an appreciated art to the masses. Your idea about a subscription service for music for film sounded outlandish at first, but I think I can see it becoming a reality. I doubt big budget films would do so, but smaller films and Netflix TV show type stuff would definitely do it. At this rate, Media Composing might not be a viable job for much longer.
sounds familiar...
Great video... I agree entirely. Just glad Apple are not charging us $$'s for Logic 11 as it's mostly 'Band In a Box' meets Toontrack EZ... only from 10 years ago.
There are sooooo many features that they could have introduced or improved to make a pro workflow more seamless, but they have decided to make it a desktop version of an iPad toy. It seems like everything they add is mostly a 'light' version 'copy' of someone else's product.
In my opinion Logic went off the rails when they fired the E Magic guys and introduced live loops. I started my DAW hobby in the 90's with Cakewalk Pro Audio 6 on a PC. I may well be going back to a PC either with Cubase, Studio 1 or perhaps even the new 'Sonar' (when they finally release it) LOL
However, thanks for your ideas on ways to use the stem splitter tool.
On AI: with AI most people can make a decent track. Some people can even make it inspiring, surprising and creative. But what got me into music was not the ability to create a final track, but rather the process where my personal feelings and imagination can be translated IRL into sound. Acoustic instruments provide great interfaces for that thing. Some tech tools can augment the experience, but mostly just interrupts the flow. With AI, I predict people will be more interested in what's "real". I think small gigs where the audience can get close to "real" musicians playing acoustic instruments will sell. The live human voice will sell. There is something in the immediacy in acoustic instruments. The total connection between my musical imagination, my fingers movements and the sounds I hear is something I've never experienced with digital products. And that's a challenge for you as a music tech developer. For example I saw some video excerpts from your string libraries and from what I can tell it sounds amazing with texture and almost a human-like aspect. But, I also see your fingers just sustaining a midi trigger while all this sound magic happens. There is this disconnection between the physical input and what I hear that has so often left me uninspired. Music is a body experience and if you can make AI products that can bring back music from the computer mouse to the human body, I'm first in line.
Stem splitter for those who find creativity in sampling is just bliss being seamlessly integrated.
Thanks for the vid, really well done and insightful ❤
Kind comment, thanks.
Maybe not entirely on topic, but I was a big hater on the Logic reverbs as well, especially Chromaverb. Until I saw Beck's Logic Pro Demo Project with his song "Colors" which is riddled with instances of Chromaverb and Space Designer. Worth taking a listen and also seeing how one can mix a song entirely using only Logic's plugins.
Great video CH. But I can't concentrate completely because of the Colossus panel behind you. It's from 'The Forbin Project', right? Brilliant.
New Spectral Layers 10 pro splits stems into 7 - vocal, drums, keyboard, guitar, bass, other, and non unmixed (any leftovers). This is the integration to choose when working with Cubase.
My guy has a $5000 mousepad. Great video.
I was relieved in one of your previous videos to hear you were finally moving to Cubase. Cubase has had this "stem splitting" tech for a couple of years now linking to another Steinberg programme called Spectral Layers which allows you to do so much more than just split mixes. Apple has just done the usual catch up and just taken the shiny bit of the tech and pretending they've created a revolution. Whilst trying so hard not to be a Logic hater, Cubase just IS so much better, it always has been. Look forward to see how you get on with Cubase/Nuendo/Spectral Layers.
Great vid Christian!
The original break from the Queen Bee's song you can hear on "Are You My Woman? (Tell Me So)" by The Chi-Lites! Amazing rhythm section overall
Chromaverb absolutely is not the worst reverb of all time. But you do, really, really, have to spend a good deal of time with it, noting how the different parameters (particularly Distance) interact with the various reverb models. And also to recognise that some of the models are non-linear and not exactly easy to get on with. But 'Room' for smaller spaces and 'Smooth Space' for larger are good starting points.
Yeah I saw an amazing tutorial just now. It sounds fantastic, but like all things Logic you really do have to dig deeper to get the right setup for each part.
As stem splitting (and all AI stuff) keeps getting more advanced it will be interesting to see if anyone uses it to create something truly interesting or original with it, sort of like how J Dilla was able to create new rhythmic concepts and combinations of samples we hadn’t heard yet. I’m not sure that the ability to remove music from film is such a game changer as I don’t feel like there’s a shortage of visual media for beginners and students to practice on. Nice to have some more options? I’m still waiting to see something that AI does that inspires me, personally. So far, it seems like it’s just providing shortcuts, cutting the need for collaborating with other people out, and lowering the barrier to entry. That’s not all bad, just not that exciting so far. I’m sure we’ll get a tsunami of mediocre music out of it. And maybe, some interesting new music if it’s not overused or if it’s purposefully misused.
Hey Christian, if you just want to split Vocals/Dialog, you can do this with standard version of Cubase in a few clicks using Spectral Layers.
Depends on which version of Cubase you have. In most cases spectral layers is a separate purchase.
I think Spectral Layers One came with Cubase in version 11 and after; and it will do Un-Mix Vocals. You do need the paid version to do more stem versions. Just wanted to share that the Vocal un-mixing is possible with the included version. Works really well for say taking a phone demo of a Vocal and Acoustic Guitar and splitting apart to get control over each element and building your tune from there. As long as it was a decent performance, this eliminates the need to go and record guide tracks. Kinda a nice timesaver if you build songs on the regular.
Ooooh thanks.
On harmonics, I tend to add one of two preamp plugins from a great little company called Fuse Audio Labs on most tracks. The VPRE-31A and 37C. Or possibly an Omega Transformer plugin from Kush Audio. Well priced, and they add some mojo as described. But willing to give Chroma-thingy a go. On the AI 'assistants', I think Drummer is brilliant for getting you close really quickly. So I can see me using the bass for that for some quick ideas...then exporting to MIDI fast. Really can't see me using the keys player at all.
Rather similar to when generative fill turned up in Photoshop, the AI things in Logic seem to belong to a different product, for people who don't really want to compose, imagine or just fiddle about. Music is only a semi-conscious activity. Automating it... negates it. The tension is between AI (backward and borrowed) vs human agency will only get starker in creative software ...
AI features are a choice. If you don’t want to use them, then you’re free to make that choice. It’s not inherently good or bad, we apply those labels to it. I’m sure great things will be created using AI tools, music included
I am using 11, I use Bitwig for creation but I am mixing a ton shows I recorded during the semester at the college I work at. seems to be working well. nice little ui changes. haven't tried the keyboardist and bassist those could be fun for writing songy type songs... u know the one