-
Notifications
You must be signed in to change notification settings - Fork 2
/
README.marsyas
55 lines (40 loc) · 2.13 KB
/
README.marsyas
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
For local audio analysis, MoodMusic uses a project called marsyas:
http://marsyasweb.appspot.com/
Marsyas allows construction of audio analysis networks. It contains
a large number of components for doing specific analysis tasks and
also contains components that can combine other components in various
ways. Any of these components can be worked on using a common
interface, so any of them can be used to analyze audio in the same way.
Marsyas is not commonly available as a binary distribution, so in order
to use it you may need to install it from source. The way I did this
was to check out the subversion repository:
svn co https://svn.code.sf.net/p/marsyas/code/trunk marsyas
Marsyas uses cmake as its build system, so you will need to install
that. Once it is installed, enter the "src" subdirectory. Next, run:
cmake ./
ccmake ./
The second command will open up an interface with which you can
configure the marsyas build. First of all, set CMAKE_INSTALL_PREFIX
to your system's normal install prefix. On most systems, this will
probably be "/usr" rather than the default of "/usr/local".
In order to use moodmusic's marsyas code, you should enable WITH_SWIG
and WITH_SWIG_PYTHON, which will enable the python bindings. You
will also probably want to enable some support for compressed audio
formats. WITH_MAD will give you support for .mp3 files, WITH_VORBIS
will give you support for Ogg Vorbis files, and WITH_GSTREAMER will
give you support for everything that GStreamer supports. GStreamer
should give you the most general support if you don't want to care
about codecs, but I have not tested it myself.
After you have instructed ccmake to generate the configuration
(using the 'c' command), exit out of it. cmake will have generated
another set of build files for your system's native build system.
On Linux systems, this will be the GNU autotools. To build marsyas,
run:
make
This will build the source. After that is complete, give yourself
extra permissions on your system and run:
make install
Now it should be installed and ready to use. To test this, you
can open up a python prompt and type:
import marsyas
If this succeeds, then you should be good.