Upvote Upvoted 16 Downvote Downvoted
Music Research Survey
posted in Off Topic
1
#1
0 Frags +

Hi!

I'm writing a new paper for a journal (specifically IEEE Journal on Evolutionary Computation), and have a survey if anyone would like to fill it out:
https://docs.google.com/forms/d/e/1FAIpQLSenjGgizcQqj_YdsdL1NNjoFO2aQbGLBEp2nZfAHHf7l1s-Kw/viewform?usp=sf_link

Your responses really do help: You may know that over a year ago I posted a couple surveys for a paper I was writing at the time (https://www.teamfortress.tv/50723/quick-music-research-survey), and thanks to your help it was accepted and presented at the IEEE Congress on Evolutionary Computation in Wellington, New Zealand last summer: https://ieeexplore.ieee.org/abstract/document/8790099
I've just successfully defended my master's thesis last tuesday on the same work, and am now writing another paper for a journal under the same as the conference.

Thank you for taking the time :) it means a lot.

Hi!

I'm writing a new paper for a journal (specifically IEEE Journal on Evolutionary Computation), and have a survey if anyone would like to fill it out:
https://docs.google.com/forms/d/e/1FAIpQLSenjGgizcQqj_YdsdL1NNjoFO2aQbGLBEp2nZfAHHf7l1s-Kw/viewform?usp=sf_link

Your responses really do help: You may know that over a year ago I posted a couple surveys for a paper I was writing at the time (https://www.teamfortress.tv/50723/quick-music-research-survey), and thanks to your help it was accepted and presented at the IEEE Congress on Evolutionary Computation in Wellington, New Zealand last summer: https://ieeexplore.ieee.org/abstract/document/8790099
I've just successfully defended my master's thesis last tuesday on the same work, and am now writing another paper for a journal under the same as the conference.

Thank you for taking the time :) it means a lot.
2
#2
6 Frags +

congrats on making it as far as you have so far and good luck with getting even more published, that's seriously cool

it almost sounds like you've fed your ai nothing but baroque music, a genre that can be notoriously complex, so it's interesting to see it trying to make something out of that style

while the music really lacks direction i think its cool that it manages to have a couple seconds here and there where there could be the beginnings of a decent idea

congrats on making it as far as you have so far and good luck with getting even more published, that's seriously cool

it almost sounds like you've fed your ai nothing but baroque music, a genre that can be notoriously complex, so it's interesting to see it trying to make something out of that style

while the music really lacks direction i think its cool that it manages to have a couple seconds here and there where there could be the beginnings of a decent idea
3
#3
2 Frags +

Thanks!
If you are interested, I didn't feed it any music. I used a genetic algorithm and built a fitness function with rules mainly from a Music-Psychology paper by David Huron.

Thanks!
If you are interested, I didn't feed it any music. I used a genetic algorithm and built a fitness function with rules mainly from [url=http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.722.3123&rep=rep1&type=pdf] a Music-Psychology paper by David Huron[/url].
4
#4
1 Frags +

Done! Well done so far and good luck furthermore :)

Done! Well done so far and good luck furthermore :)
5
#5
0 Frags +

will help out a fellow spark

will help out a fellow spark
6
#6
3 Frags +

these beats are cold

these beats are cold
7
#7
1 Frags +

Turns out I made an error on the survey.. I updated it with a new one. Sorry everybody
https://docs.google.com/forms/d/e/1FAIpQLSfS9QkB9TJd2H7hKgM_yNZHrymvTDc3cZa29Al2bG-Fi3XWLg/viewform?usp=sf_link

Turns out I made an error on the survey.. I updated it with a new one. Sorry everybody
https://docs.google.com/forms/d/e/1FAIpQLSfS9QkB9TJd2H7hKgM_yNZHrymvTDc3cZa29Al2bG-Fi3XWLg/viewform?usp=sf_link
8
#8
1 Frags +

Its nbd! lol, the new ones sound so much better :o

Its nbd! lol, the new ones sound so much better :o
9
#9
0 Frags +

how long did it take to train for these samples?

how long did it take to train for these samples?
10
#10
8 Frags +

ahhh these questions were so hard i hope i pass lol

ahhh these questions were so hard i hope i pass lol
11
#11
1 Frags +

Damn this is really cool I'd be interested to read the paper once you're finished. How come the different snippets would alternate on what synth keyboard they're using or whatever, was that just to make it easier to distinguish them?

Damn this is really cool I'd be interested to read the paper once you're finished. How come the different snippets would alternate on what synth keyboard they're using or whatever, was that just to make it easier to distinguish them?
12
#12
1 Frags +
Bob_MarleyIts nbd! lol, the new ones sound so much better :o

Funnily enough, half of these are the old ones.

tuahow long did it take to train for these samples?

I'm not using a learning algorithm so there was no training. Each one on this survey takes a couple minutes to generate.

bearodactylDamn this is really cool I'd be interested to read the paper once you're finished. How come the different snippets would alternate on what synth keyboard they're using or whatever, was that just to make it easier to distinguish them?

That was mostly my fault, I used different instruments for each without noticing. I do this to avoid having bias in selecting which compositions I use (I generate 10 and randomly choose 5).
One used harpsichord and the other piano. That might actually confound results :x whoops

[quote=Bob_Marley]Its nbd! lol, the new ones sound so much better :o[/quote]
Funnily enough, half of these are the old ones.

[quote=tua]how long did it take to train for these samples?[/quote]
I'm not using a learning algorithm so there was no training. Each one on this survey takes a couple minutes to generate.

[quote=bearodactyl]Damn this is really cool I'd be interested to read the paper once you're finished. How come the different snippets would alternate on what synth keyboard they're using or whatever, was that just to make it easier to distinguish them?[/quote]
That was mostly my fault, I used different instruments for each without noticing. I do this to avoid having bias in selecting which compositions I use (I generate 10 and randomly choose 5).
One used harpsichord and the other piano. That might actually confound results :x whoops
13
#13
2 Frags +

I posted yet another one without this confound. Sorry, yet again :x
https://docs.google.com/forms/d/e/1FAIpQLSenjGgizcQqj_YdsdL1NNjoFO2aQbGLBEp2nZfAHHf7l1s-Kw/viewform?usp=sf_link

I posted yet another one without this confound. Sorry, yet again :x
https://docs.google.com/forms/d/e/1FAIpQLSenjGgizcQqj_YdsdL1NNjoFO2aQbGLBEp2nZfAHHf7l1s-Kw/viewform?usp=sf_link
Please sign in through STEAM to post a comment.