The Future of Musical Instruments.

Xurtio · 02-21-2015, 07:21 PM

Quote:

Originally Posted by Chula Vista

Dude. Really?

He's not just playing notes. He's feeding off of the whole thing. Watch his face throughout the song. It's maybe 50% notes and 50% emotionally feeding off of the note he just played. A computer will never be able to emotionally improv like Hedges does in that clip.

Ok, but in terms of the physics, those nuances all come down to the shape of the attack, continual pitch shifts for bends, shape of the release and the sustain. It can all be described with one time-dependent variable.

With the right rule system or even supervised training on Hedges himself, an algorithm could generate such waveforms.

GuitarBizarre · 02-22-2015, 10:43 AM

In this thread: Half of the people don't understand how computers work, and are falling back on "BUT EMOTION MAN! COMPUTERS CAN'T FEEL EMOTION!"

Nevermind that we've been considering the possibility of intelligent machines for decades and getting closer in a very constant way.

The other half are saying "Uh, you do realise the moment we establish a way to quantify emotion in a logical system, computers will be able to understand emotion and replicate it? Literally the drawback is that because we don't have a flawless understanding of brain chemistry and response to external stimuli, current approaches are limited to responding to broader and less specific inputs, but eventually we will hit the point where those approaches are refined and become viable, or we will hit the point where constructing a computerised simulation of brain activity during this act can be done in synchronicity with the act of running a musical output?"

It is entirely conceivable - in fact, likely - that once we understand the operation of the brain to the point we can simulate it's internal workings, we will actually be able to directly synthesise music from thoughts, bypassing the need for using a physical instrument or digital synthesizer. We, or our hypothetical brain simulation, will literally be able to THINK "I want a sound like this", and get it. In realtime. We will be able to THINK a wah pedal or a timpani being hit "just so" and get those timbre immediately without intervention by a third party system or an object.

Once that level of advancement arrives, or we even get CLOSE to it, computers will already be very much capable of emulating, imitating, and creating any musical form, including any part of that form which is improvisational or a response to outside stimulus.