Music and Machine Learning (Google I/O'19)

New technologies have fundamentally changed the way we make and experience music. In this session Claire Evans, artist, author and one half of the pop duo YACHT talks about deep learning as a tool in their creative process. Their new album explores Google AI’s research project, Magenta, an open-source music-making package using machine learning models.
Watch more #io19 here:
Inspiration at Google I/O 2019 Playlist → goo.gle/2LkBwCF
Google I/O 2019 All Sessions Playlist → goo.gle/io19al...
Learn more on the I/O Website → google.com/io
Subscribe to the Google Developers Channel → goo.gle/develo...
Get started at → developers.goo...
Speaker(s): Claire Evans, Adam Roberts, Jesse Engel
T1AFC0 event: Google I/O 2019; re_ty: Publish; fullname: Claire Evans, Adam Roberts, Jesse Engel;

[@rezik007]

Being a drummer and programmer...I'm totally terrified :D

[@dansanfridsson8991]

8

[@7racker]

Totally terrified but smiling after you say it. Can you expand on that?

[@malika8934]

But look what she mentioned - " It`s not that we just click a button "give me a song" but that you need to have a vision, an interpretation", that`s the main thing in art, in my opinion. The tools might change but still this kind of artistic vision and taste - how to use something, that output gives you is required to make it useful. This drum loop you get on a click isn`t something we don`t have already - there`s tons and millions of ready to use drum loops, I think that there`s so many that we already covered all the variations. And this plugin is just something that ease you the effort of finding a proper fitting loop, it just creates one. Anyways what magenta does are amazing tools, really cool :D

[@JesseScott2016]

I love the fact the standalone Magenta Studio tools are free and open source! Providing great tools to people like me that want to experiment with music! Also the flaming lips performance was great

[@wakinguponjupiter7505]

Great Job! Thanks for exploring YACHT, I'm a big fan.

[@crazydrifter13]

I find artists to be a little bit weird

[@charlesleflamand]

:))

[@laconsciencedetre2760]

Hello,
Please I am looking for important information for me.
Is there a way to have coconet (bach) on your computer so that you can use it on a daily basis?
Thank you for your attention
Oliver D.
Composer

[@arthurduarte5212]

28:30 Sweeney Todd starts a beef with Sweet Dreams vocalist

[@headfonic7750]

Great ! It looks really expert! Keep up the great work!

[@oumood]

Holly shiiiit

[@JamesLewis2]

The part that shows up as [INAUDIBLE] in the subtitles is "mondegreen".

[@mohamedazizabidi84]

that's my idea

[@hiddenhandconsultants1619]

Apple logic did this years ago

[@quosswimblik4489]

So you have your algorithm identifying things but it only can identify stuff its familiar with in familiar scenes. Whats missing if I see a pink sheep on a stair case my mind adapts but a computer will likely relate it to something it is familiar with and identify that there's a pile of flowers on a stair case. So in short AI is today good at deep learning with things it can get familiar with but is not good at adaptive learning like a human or even an ant colony can be.
So where do you start well I would suggest a robot like ant colony where you challenge the ant colony with things unfamiliar to it to test new systems abilities to adapt or that sort of thing. Imagine that Ai in the future ran more of our lives but wasn't very adaptive this could lead to some small or big problems as new challenges arose. It also gets one thinking about the nature of consciousness is consciousness simply a mechanism that allows life to adapt to new challenges.

[@secretmeeting4886]

AI actually works far more like adaptive learning than that. Machine learning uses a dataset to produce a model, and then a human feeds it unfamiliar data with which it updates the model. This is essentially what humans do, no? Hence we also have biases and we also aggregate people and other objects in the world based on suppositions that are derived from our experience. The limitations of AI currently seem to me to be computational and data statistical, rather than a fundamental ontological failure. Also, the ant colony idea seems at least superficially similar to neural network based AI.

[@quosswimblik4489]

@@secretmeeting4886 when I look at AI using augmentation i feel a bit like i'm peering into a five year old's mind. All these weird neural augmentations together build a perception or picture in your mind at any given moment of consciousness and the bit that really alerts our perceptual field of view.

[@AnthonyBatt]

"Every word is made word." Wow Claire thanks for that clairty

[@expertineveryarea.8950]

its very intuitively

[@vindastris4982]

26:00 Играет на фруктах XD

[@stevewright9195]

Nice!