TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Music Language Modeling with Recurrent Neural Networks

100 pointsby yoavzabout 9 years ago

8 comments

iammyIPabout 9 years ago
As usual* the result sounds awfully unstructured and unenjoyable and could be aswell be achieved by some random walk through a musical scale, since if you put some basic music theory in program form, you can get these small harmonic structures pretty easily. (*since projects like this seem to pop up every couple of months)<p>Like other people already mentioned, the part that usually gets neglected is the overarching dramatic structure of a musical piece. Compare a complete shakespeare piece to a pile of randomly thrown together half-sentences.<p>I don&#x27;t fully understand the fascination of music generation with some ai-neural-learning buzzword bingo technique that always gets kickstarted by dumb-force-analysing a human made music corpus to achieve it.<p>What in a musical sense is much more interesting is to generate _new_ music that cannot be composed by a human, and cannot be played by a human. That&#x27;s playing to the strength of the machines. Sonification of large datasets, sonification of function behaviour. Sonification of the binary world, that&#x27;s so different to ours. This is much more interesting than the 10th failed emulation of a simple folk song.<p>Nevertheless, as a students piece about programming neural networks, it&#x27;s certainly ok, the presentation is nice, but the result is uninspiring, like building a car tire out of bananas, just because it&#x27;s possible. Just let the folk songs belong to the actual folk.<p>As a side note: what would happen if the result were millions of super nice catchy folk tunes on a button press? Would it be the end of pop music as we know it? Maybe i redact my opinion.
评论 #11406524 未加载
评论 #11409970 未加载
评论 #11406548 未加载
albertzeyerabout 9 years ago
I&#x27;m quite sure I have seen other attempts to generate music with RNNs recently, although I don&#x27;t remember exactly anymore. You don&#x27;t cite that many references to other approaches, only the one from Boulanger-Lewandowski from 2012.<p>I did a quick search, and I probably miss a lot, but I found these:<p><a href="http:&#x2F;&#x2F;papers.nips.cc&#x2F;paper&#x2F;5655-deep-temporal-sigmoid-belief-networks-for-sequence-modeling" rel="nofollow">http:&#x2F;&#x2F;papers.nips.cc&#x2F;paper&#x2F;5655-deep-temporal-sigmoid-belie...</a><p><a href="http:&#x2F;&#x2F;dl.acm.org&#x2F;citation.cfm?id=2806383" rel="nofollow">http:&#x2F;&#x2F;dl.acm.org&#x2F;citation.cfm?id=2806383</a><p><a href="http:&#x2F;&#x2F;gitxiv.com&#x2F;posts&#x2F;WEoQCj8hxHz6vPxe6&#x2F;gruv-algorithmic-music-generation-using-recurrent-neural" rel="nofollow">http:&#x2F;&#x2F;gitxiv.com&#x2F;posts&#x2F;WEoQCj8hxHz6vPxe6&#x2F;gruv-algorithmic-m...</a> <a href="https:&#x2F;&#x2F;github.com&#x2F;MattVitelli&#x2F;GRUV" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;MattVitelli&#x2F;GRUV</a><p><a href="http:&#x2F;&#x2F;www.hexahedria.com&#x2F;2015&#x2F;08&#x2F;03&#x2F;composing-music-with-recurrent-neural-networks&#x2F;" rel="nofollow">http:&#x2F;&#x2F;www.hexahedria.com&#x2F;2015&#x2F;08&#x2F;03&#x2F;composing-music-with-re...</a> <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=10028878" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=10028878</a><p>I haven&#x27;t really looked into any of these, so I&#x27;m not sure about the differences. But it would be good if you cite some relevant other works and point out the differences.
romanivabout 9 years ago
In my opinion, anyone who works on music generation should take a look at Karma (<a href="http:&#x2F;&#x2F;karma-lab.com&#x2F;" rel="nofollow">http:&#x2F;&#x2F;karma-lab.com&#x2F;</a>) for a baseline of what can be achieved by simple math and plain old programming. Probably not particularly interesting from programmer&#x27;s perspective (it&#x27;s closed-source and to the best of my knowledge doesn&#x27;t use anything fancy), but the end results are spectacular and used in real music.
评论 #11407489 未加载
评论 #11407647 未加载
评论 #11407337 未加载
akleinabout 9 years ago
I would love to see a real-time bebop improvisation generator in the style of Charlie Parker, Sonny Rollins, Bill Evans, et. al. - bebop is definitely a musical (jazz) language that I bet would be well suited to RNNs.
评论 #11405774 未加载
评论 #11405742 未加载
cel1neabout 9 years ago
You could feed Lilypond [0] files into the network. You might gain more long-term structure that way.<p>It looks like this, you can do repeats and everything else:<p><pre><code> \new Voice \with { \consists &quot;Ambitus_engraver&quot; } \relative c&#x27; { \voiceTwo es4 f g as b1 } </code></pre> [0] <a href="http:&#x2F;&#x2F;lilypond.org&#x2F;" rel="nofollow">http:&#x2F;&#x2F;lilypond.org&#x2F;</a>
评论 #11407726 未加载
varelseabout 9 years ago
Both of the pieces at the top of the article sound like off-key broken record renditions of the main refrain of &quot;Jesu, Joy of Man&#x27;s Desiring&quot; to me. It&#x27;s like the RNN cannot hold enough state to express the structure of a real musical piece and it just emits riffs here and there of main themes from its training set.<p>What would be somewhat impressive is if it spontaneously figured out the note sequence I hear from observing its re-expression in bits and pieces from various jigs and folk pieces in its training set, kind of like this:<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=XPLp_gInC-o" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=XPLp_gInC-o</a>
评论 #11405335 未加载
studentrobabout 9 years ago
Cool. Anyone have an opinion on &quot;state of the art&quot; for music generation? I realize this is entirely subjective. This one sounds pretty interesting! It&#x27;d be awesome to get something like this on a top 10 list and start influencing man made music. We can&#x27;t be so far off from that. The kids these days love techno and that is easily synthesized relative to music with original lyrics and voices.
评论 #11405469 未加载
评论 #11408193 未加载
6stringmercabout 9 years ago
Fascinating piece of research and the details in the write-up managed to click mostly even though I know it&#x27;s a level far above my head. Well done and glad to have come across it.