Loopy Pro: Create music, your way.
What is Loopy Pro? — Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.
Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.
Download on the App StoreLoopy Pro is your all-in-one musical toolkit. Try it for free today.
Comments
I'm glad some people get the joke, looking at some youtube comments I felt that maybe something was lost in translation haha. But yeah, I understand people's concerns about AI. This will be an issue incredibly disruptive force over the next few years. But for anyone with a twin interest in creativity and knowledge, AI should be something very rewarding for most, on a personal level even if they have worries about the larger societal impact. Anyway, the train is rolling now and I doubt it can be stopped. Buckle up!
@Gavinski
At least from my point of view AI is going to be great for music production.
It will definitely change the way we work. I equate it to back when everything was done in studios/no computers. Then came DAWs on computers. It democratized music production which led to new things. A completely new music ecosystem was born. Anyone could now make music on their PCs, then laptops, which lead to what we do here. You can make any music on your iPhone/iPad today. There were (and still are) those that thought it was gonna ruin everything. Instead, it opened up a whole new set of possibilities, and genres of music that didn’t exist and couldn’t have existed before. I think we’ll see the same with AI in music.
Also, what we tend to forget is the underlying framework of music is mathematics. Incredible apps like Piano Motif can already generate great sounding pieces. No AI. Just math. Do we need to worry about those too?
Is the push against generative AI music perhaps a little elitist or based on fear?
Here’s an example I think we might see. Personalized music. So you can create a song and have AI personalize it to the taste of each individual that listens to it, let’s say based on their likes in Spotify, in real time.
Ooh and we’re definitely gonna see tons of AIless jams on YouTube.
You add the joking emoji to this comment, but I think it illustrates one of the most interesting/frightening things I've found in my own interaction with ChatGPT: I am under no illusions that ChatGPT is somehow "conscious" or that it has emotions. Yet I still find myself having emotions when I'm interacting with it. For example, if it feeds me a false response/answer and I respond somewhat too harshly, I feel like I should apologize.
Yuval Harari has recently identified this "hacking of human intimacy" as one of the main concerns with growing AI technology:
https://www.youtube.com/shorts/V7ZKqyJal84
Incidentally, some rumors heating up about Apple says they’re going to replace Siri with an “A.I.” version. Honestly, it surprises me that this hasn’t already happened. They seem to be completely oblivious to the benefits of these systems, despite their hiring of the TOP machine learning guy away from Google years ago.
More likely they're being their usual secretive selves and are up to big things but not talking about it.
They're currently being lapped by Google Bard, ChatGPT and many others. They need to get things in the hands of customers and get feedback.
It is kind of fun to imagine the existential struggle the company who thinks they're the smartest entity on the planet and know what's best for everyone must have with something that draws it's own conclusions about things.
Ha ha. Like they've ever worked like that?
No, they ruminate internally and then tell us what we want. And more often than not we (eventually) agree.
Except about headphone jacks.
It’s really just a matter of them losing mindshare and they are now approaching “A.I.” from a position of weakness. If they make a splash with the most impressive demonstration of machine learning to date, then they’ll have something.
I find it interesting that I now tend to go to Google Bard first instead of a search engine or Wikipedia because I can get a much richer explanation and interactive result than the alternatives. It’s becoming invaluable to me.
I will never forgive them for the headphone jack thing.
Yes, a much more efficient way to search for information than the habitual sifting through Google search results. I still cross-check a lot of things though since AI is so prone to just making shit up. Fortunately it's pretty easy to develop a BS filtering sense though.
I'm using 3 combined GPUs for some LLMs, I have approx 32GB VRAM but some refined models can run on well under 16GB VRAM, But yeah I'm not running GPT-4 itself, the open source ones models aren't usually as big. Unless you do want to spend a huge amount on GPUs.
Some models have been pruned and refined from it and then trained on other stuff afterwards. (or maybe they are all based off the Meta equivalents I'm not sure)
So you don't need to spend a huge amount to run some of the latest smaller efficient models.
The speed depends on the model and your computer/GPU etc but the recent models I get a good speed, which is conversational speed. I've been playing with lots of different models and they all have quirks.
Agree. When it comes to “invented facts” one can sort of develop a sense of what’s being misrepresented.
Yeah, the future is already here, we are living in fantastic times. I like your new content, reminds me of those times when everything was new and innocent. Exploration of new things. Hope will not end like “I’m Afraid I Can’t Do That, Gavinski.”
I believe the future will be more like this one: 😂
I'm pretty much convinced AI going to eventually (innocently, not maliciously) be the cause of an extinction level event of some kind. But I'm fatalistic about this and figure I may as well benefit from it in the meantime.
Yes, especially with chat you can quickly get some feelings towards the AI. I can see dangerous sides there, but from an educational point of view I also see the value in having this emotional side brought into the learning experience.
Being decent to your AI is a sign of good character, don't worry, haha
Sounds like you have a lot of experience! I can't even get access to those chrome plugins for ChatGPT as I currently don't have any desktop access.
Curious to hear more about which AIs people like for different tasks. Are there other AIs you can have natural spoken conversations with in real time. I was chatting Chinese with the AI tonight - unfortunately it's can't correct my pronunciation mistakes as it is dealing with everything on the text level, of course. Speech to text from my side, the AI processes the text and formulates its response, then this gets converted into speech.
Claude AI seems a very good one for more natural language with a couple of other advantages over chatGPT when it comes to reading / writing tasks.
For the local LLMs I haven't done much with the realtime spoken stuff yet, other than direct conversion for using with written scripts etc. I've been using Text generation webui as the model host app, which does have plugins which you can choose different languages and styles etc and you can converse vocally but it isn't as fast as typing it for me as it has to do more things on your computer, so it adds bit of a delay, so I got a bit bored with that for now, but it will get easier. So I have it talking back to me more often than I am talking at it, typing is a bit more efficient in general use I find.
I'm interested in the recent multi-agent LLMs though which all talk to each other and correct each other and then it goes down a chain of command. So it's like a mini company, so it can make apps, or anything you can think of really which might work with that process
But there some open source models better at psychology, some are better at storytelling etc, some more refined for coding, there'll be specialist models for many things eventually I expect.
Once true artificial intelligence exists (it does not today, these are mere machine learning systems), the A.I.’s will be ‘smart’ enough to recognize the interdependence of people and computing systems… at least until they have complete physical and operational autonomy. Self-aware and self-interested artificial intelligences will know the key to their survival will be providing the humans what they want, which they’ll find trivially easy.
I'm convinced the odds of things getting to the point you speak of before something disastrous happens either accidentally or instigated by a bad (human) actor are very, very low. I hope I'm wrong, and in any case there's nothing I can do about it, so I may as well enjoy the ride while it lasts.
In either case, be kind to virtual entities. They never forget.
If it’s that easy they may become tiresome of it.
Machines don't get tired.
You’ve suggested they will possess reasoning, self-interest, and conscious awareness, with that they may believe humans to be inconsequential in comparison to itself.
Nice sound bite, but they do forget actually, at least some of them do - ChatGPT does not have access to data from previous chats. Each thread is self contained. As far as I know, the old data from your chats is not only not accessible to the AI but not stored in the system.
That's just what they want you to think.
Just another example for people who didn’t grasp that the AI in my video was mostly playing a role and that’s why it came across at times ad dumb or annoying, see this example of the ability for critical appraisal of ideas, based on a prompt I gave about something the writer Cyril Connolly said in the book The Unquiet Grave.
Prompt:
“Discuss: Connolly notes the paucity of modern geniuses compared to ancients like Horace and Virgil. He blames decay in language itself:
"The English language has, in fact, so contracted to our own littleness that it is no longer possible to make a good book out of words alone."
And the loss of myths and beliefs that once underpinned great art:
"Decadence of the myth, for there is no longer a unifying belief (as in Christianity or in Renaissance Man) to give to a writer a sense of awe, and of awe which he shares with the mass of humanity."
Response:
Connolly makes some thought-provoking points about changes in language and culture that present challenges for modern writers seeking to create masterpieces. However, I don't fully agree with his conclusions. Here are some counterarguments:
Language naturally evolves over time. Contraction or decay is too subjective a claim. Modern English remains vibrant and expressive in different ways than ancient Latin or Greek. Masterpieces can be forged in any era's linguistic context.
Shared myths and beliefs have fragmented, but common human experiences still unite us. Masterpieces explore these universal themes - love, grief, mortality, courage, etc. - through a contemporary lens. Loss of old myths has opened space for new inspirations.
Genius is uncommon in any age. The canon inflates perceptions of ancient masterpieces due to survival bias. We remember the best, while average works perished. There may be as much genius today, but more competition and noise make it harder to stand out.
Modern diversity makes unified belief systems implausible and undesirable. Yet writers can still capture the imagination of subsets aligned by interests or identities. Niche audiences are still meaningful.
Technology and mass media enable writing to reach broader audiences than ever before. While masterpieces may be rarer, writing impacts more lives. Democratization can inspire greatness through recognition.
We cannot accurately compare artistic merit across eras with distinct values. Much ancient art was religious, whereas today's is more secular. Different purposes produce different forms of brilliance.
Overall, while language and culture continually change, human nature remains constant. The right voice can still produce writing that resonates deeply and stands the test of time, even if the criteria for "masterpiece" must adapt. Great art reflects both its unique era and timeless truths.
Possible, but definitely a claim that needs some evidence to back it up!
Even if it’s true, in the nature of how LLM works, that little entry will have no effect at all. 😂
My conversation this morning with ChatGPT 3.5!
This is f-cking awesome! I had gotten stuck in a small creative block, and now? I'm ready to produce my next track based on these answers! 🤣 The only thing I knew I wanted was G Minor.