Saturday, 10 June 2017

Neural Translation of Musical Style

quote [ Currently, generated performances lack dynamics and sound monotonous. This opens doors for adding musical style to the compositions! This all leads to one interesting question: Can a machine perform like a human? ]
[SFW] [music] [+6 Interesting]
[by arrowhen@6:18pmGMT]

Comments

lilmookieesquire said @ 9:47pm GMT on 10th Jun
Don't they already have things like this where the program fills in the background music?

The Baby llama in red pajamas YouTube post had a comment from someone who suggested they used a program like that to fill in the background (and it was pretty good)
spaceloaf said[1] @ 10:17pm GMT on 10th Jun [Score:1 Interesting]
I don't think that's what this is.

This is specifically adding dynamics to a pre-programmed track of notes to make it sound as if a human was playing them (meaning not every note is hit with the exact same intensity; certain notes are emphasized to reinforce the melody/musical phrasing).

In other words the computer didn't generate the music; it is trying to simulate a person playing an already existing piece of music. What makes it interesting is that the computer must "understand" the way humans perceive music (to some basic level at least) in order to do this.
lilmookieesquire said @ 9:47pm GMT on 10th Jun

Post a comment
[note: if you are replying to a specific comment, then click the reply link on that comment instead]

You must be logged in to comment on posts.



Posts of Import
Karma
SE v2 Closed BETA
First Post
Subscriptions and Things

Karma Rankings
ScoobySnacks
HoZay
Paracetamol
lilmookieesquire
Ankylosaur