|
c@0
|
1 %% Multichannel audio source separation evaluation framework (MASSEF)
|
|
c@0
|
2 %
|
|
c@0
|
3 %% Overview
|
|
c@0
|
4 %
|
|
c@0
|
5 % The multichannel audio source separation evaluation framework is designed
|
|
c@0
|
6 % to faciliate the development and evaluation of audio source separation
|
|
c@0
|
7 % algorithms. The framework generates the mixture(s), calculates the ideal
|
|
c@0
|
8 % binary and ratio masks, provides the mixture(s) to the separation
|
|
c@0
|
9 % algorithm(s), and evaluates the outputs of the separation algorithm(s).
|
|
c@0
|
10 % The framework also evaluates the ideal masks for the purposes of
|
|
c@0
|
11 % comparison.
|
|
c@0
|
12 %
|
|
c@0
|
13 % Sources may have any number of channels; the framework evaluates each
|
|
c@0
|
14 % channel. The use of iosr.bss.mixture objects facilitate the evaluation of
|
|
c@0
|
15 % spatialised mixtures (e.g. binaural).
|
|
c@0
|
16 %
|
|
c@0
|
17 % The framework comprises two main classes:
|
|
c@0
|
18 %
|
|
c@0
|
19 % * <help_MASSEF.html MASSEF>—multichannel audio source separation
|
|
c@0
|
20 % evaluation framework class.
|
|
c@0
|
21 % * <help_MASSEFresults.html MASSEFresults>—the results class used by
|
|
c@0
|
22 % MASSEF.
|
|
c@0
|
23 %
|
|
c@0
|
24 % <help_Example.html View an example of the use of these classes.>
|
|
c@0
|
25 %
|
|
c@0
|
26 % MASSEF can be used to evaluate and compare binaural separation algorithms,
|
|
c@0
|
27 % provided that the algorithms conform to the <help_Separators.html
|
|
c@0
|
28 % required format>.
|
|
c@0
|
29 %
|
|
c@0
|
30 %% Installation
|
|
c@0
|
31 %
|
|
c@0
|
32 % Before using the framework, you need to download and install its
|
|
c@0
|
33 % dependencies. This can be done automatically using the
|
|
c@0
|
34 % <help_MASSEF_install.html MASSEF.install> method. Before each session,
|
|
c@0
|
35 % MASSEF needs to be started using the <help_MASSEF_start.html
|
|
c@0
|
36 % MASSEF.start> method.
|
|
c@0
|
37 %
|
|
c@0
|
38 %% Usage
|
|
c@0
|
39 %
|
|
c@0
|
40 % The framework can be run in two ways:
|
|
c@0
|
41 %
|
|
c@0
|
42 % # by providing |iosr.bss.mixture| objects and separation algorithms, or
|
|
c@0
|
43 % # by providing estimate and true source wav files.
|
|
c@0
|
44 %
|
|
c@0
|
45 % If 1), the framework generates the mixture(s), calculates the ideal
|
|
c@0
|
46 % binary and ratio masks, provides the mixture(s) to the separation
|
|
c@0
|
47 % algorithm(s), and evaluates the outputs of the separation algorithm(s).
|
|
c@0
|
48 % The framework also evaluates: the ideal masks for the purposes of
|
|
c@0
|
49 % comparison, and any azimuth/elevation estimates returned by the
|
|
c@0
|
50 % algorithm. Use the <help_MASSEF_execute.html execute> method to operate
|
|
c@0
|
51 % in this mode.
|
|
c@0
|
52 %
|
|
c@0
|
53 % If 2), the framework evaluates only the supplied estimate(s) using
|
|
c@0
|
54 % signal-related metrics. Use the <help_MASSEF_evaluate.html evaluate>
|
|
c@0
|
55 % method to operate in this mode.
|
|
c@0
|
56 %
|
|
c@0
|
57 % <html>
|
|
c@0
|
58 % <hr>
|
|
c@0
|
59 % <p>Copyright © 2016 <a href="http://www.surrey.ac.uk">University of Surrey</a><br>
|
|
c@0
|
60 % <a href="http://iosr.uk">Institute of Sound Recording, University of Surrey, UK</a></p>
|
|
c@0
|
61 % </html>
|
