Musical Machine Learning - Revision history

Nsbradford: /* Musical Machine Learning */

2016-05-05T05:53:21Z

‎Musical Machine Learning

2016-05-05T05:52:23Z

‎Musical Machine Learning

2016-05-05T05:43:32Z

‎TODO List

2016-05-05T05:38:31Z

‎Musical Machine Learning

2016-05-05T05:37:28Z

2016-05-05T04:09:08Z

2016-05-05T04:03:32Z

2016-04-21T16:31:42Z

2016-04-20T18:49:41Z

2016-04-20T18:39:14Z

@@ Line 1: / Line 1: @@
 =Musical Machine Learning=
-[[File:Confutatis.png|200px]]
+[[File:Confutatis.png|600px]]
 ==Overview==

@@ Line 8: / Line 8: @@
 *Large additions
-**Cluster an entire song in 1-s segements, then use a Gaussian KDE to smooth out classification. This can then be used to actually mark "segments" of a song
+**Cluster an entire song in 1-s segments, then use a Gaussian KDE to smooth out classification. This can then be used to actually mark "segments" of a song
 **Train a massive Deep Neural Net to try to automatically distinguish between parts
 **Composition
@@ Line 24: / Line 24: @@
 **Optimize song loading times (store in database? alternative form?)
 **Add option of multiple sections (bridge?)
 ==Help Connecting to Repository==

← Older revision		Revision as of 05:37, 5 May 2016
Line 4:		Line 4:

	==Overview==		==Overview==
		+
		+	This project created a foundation for future work by WPI students in music information retrieval and machine learning. A Python system was first constructed to extract variable-length features from audio files. The problem of determining song structure was then approached with both supervised and unsupervised learning algorithms, resulting in a novel method for automated structure analysis. Finally, the groundwork was laid for the use of deep neural networks for musical composition.

	==TODO List==		==TODO List==

@@ Line 1: / Line 1: @@
 =Musical Machine Learning=
+[[File:LoopBuddy1.0.PNG|200px]]
 ==Overview==
@@ Line 7: / Line 7: @@
 ==TODO List==
-*Use SVM with linear kernel (instead of RBF) as a better approximation of "true" clustering
+*Large additions
-*Cluster an entire song in 1-s segements, then use a Gaussian KDE to smooth out classification. This can then be used to actually mark "segments" of a song
+**Cluster an entire song in 1-s segements, then use a Gaussian KDE to smooth out classification. This can then be used to actually mark "segments" of a song
-*Create website where people can upload a MIDI file, and then listen to RNN improvise over it
+**Train a massive Deep Neural Net to try to automatically distinguish between parts
-*IPython notebook demo
+**Composition
-*Verse/Chorus system:
+***Create a LSTM recurrent neural net to learn from MIDI input
 **Find additional features other than spectral centroid and zero-crossing rate
 ***Play with how the features are generated and averaged
 **Include outlier detection in the data preprocessing stage
@@ Line 18: / Line 23: @@
 **Optimize the different classifiers
 **Optimize song loading times (store in database? alternative form?)
 **Add option of multiple sections (bridge?)
-*Expansions:
-**Train a massive Deep Neural Net to try to automatically distinguish between parts
 ==Help Connecting to Repository==

@@ Line 25: / Line 25: @@
 <br>
 <br>
-[[Category: Advisor:Manzo]]
+[[Category: Advisor:Manzo]][[Category:Interactive Systems]]

← Older revision		Revision as of 18:39, 20 April 2016
Line 1:		Line 1:
	=Musical Machine Learning=		=Musical Machine Learning=
		+
		+	==Other Resources==
		+	*'''All Unity Assets'''
		+	**[http://www.vjmanzo.com/wpi/Unity/Asset%20Store.html All Unity Assets] See library, then contact [[User:Vjmanzo\|Manzo]] for specific assets
		+	*'''Sound FX Library'''
		+	**smb://research.wpi.edu/musictechnology (on _All Student Resources shared folder) [http://wiki.wpi.edu/helpdesk/Network_Drive_Mapping Connecting to the Shared Folder]
		+	*'''IQP Papers'''
		+	* [https://www.wpi.edu/Pubs/E-project/Available/E-project-042815-193112/ The Loft]
		+	*Fuse on the Lab Computers
		+	**Adobe Fuse is installed [[Media:Fuselab.png \|here]] on two computers in the corner of the Multimedia lab, over near the tech suite in the Library.
		+	<br>
		+	<br>
		+	<br>
		+	<br>
		+	<br>
		+
		+	=WPI Student Contributors=
		+	==2016==