PH_Pooled Featrues Classification MIREX 2011 Submission

Abstract

1. Principal Mel-Spectrum
   Components (Feature)


2. Temporal Pooling
   Functions (Model)


3. Single Hidden Layer Neural Network, thus Multi-layer Perceptron
   (Classifier)

Audio Preprocessing

Feature: PMSC (Principal Mel-Spectrum
Components)

1. Original Data:
    30s, 22.05KHz, mono, wav


2. Process
   Steps:


1. DFT (spectral
   domain)
   we compute DFTs over windows of 1024
   samples on audio at 22.05 KHz (i.e. roughly 46ms) with a frame step
   of 512
   samples.


2. Mel-Compression
   we
   run the spectral amplitudes through a set of 256
   mel-scaled triangular filters to abtain a set of spectral energy
   bands.


3. Principal Component
   analysis whitening (PCA whitening)
   we compute the principal components of
   a random sub-sample of training set. In order to obtain features with
   unitary variance, we multiply(乘以) each component by the inverse square of
   its eigenvalue(特征值平方的倒数). ---- PCA whitening.

Model

PFC (Pooled Features
Classifier)

1. Pooling Operation
   the model applies a given set of pooling functions
   (how many?) to the PMSC features, and sends the pooled features to a
   classifier(MLP, with hidden layer of 2000 units, sigmoid activation, L2 weight
   decay and cross-entropy cost).


2. Classify
   each pooling window is considered as a training example for
   the classifier, and average the predictions of the classifier over all the
   windows of a given clip to obtain the final classification (what is the
   rule?).

Tasks

1. Classification (train/test task)
   the MLP outputs an affinity prediction
   for each class (pooling functions tread each pooling window as a training
   example).


2. Tagging


1. Affinity
   the
   affinity scores for a song is
   thus directly the output of the MLP.


2. Binary Classification
   choose the threshold that optimizes the
   F1-score on the validation set.

Tools

1. Theano: Theano is
   a numerical computation library for Python. In
   Theano, computations are expressed using a NumPy-like
   syntax and compiled to
   run efficiently on either CPU or GPU architectures.

   来源： <http://en.wikipedia.org/wiki/Theano_(software)>

来自为知笔记(Wiz)

PH_Pooled Featrues Classification MIREX 2011 Submission