On Dimecres 18 Abril 2007, jn jn wrote: > Hello, > > I'm new to CLAM, and my interest on audio has been increasing in the last > months, mainly because of my interest in MPEG-7 and in audio > fingerprinting. And so I was wondering if CLAM has been coded taking in > care what is written in the MPEG-7 audio standard and also if it supports > audio fingerprinting (by itself).
One of the examples on the Annotator extracts most of the Mpeg7 low level descriptors. Most of the Mpeg7 descriptors has been review against the standard and some of them are not what the are intendeded (notably Spectral Kurtosis and Skew). You'll find an accurate documentation on Anyway when they fail to implement the standard it is clearly noticed on the doxygen documetation. We didn't remove them because even thought they are not the standard, they has been proven useful in some contexts. About fingerprinting, i would suggest to use musicbrainz library. If you want to implement such a system in CLAM you are wellcome. Anyway if you explain better your goals maybe i could answer you more accuratelly. You can browse what do we have by checking some of those classes: http://clam.iua.upf.edu/doc/CLAM-devel-doxygen/classCLAM_1_1AudioDescriptors.html http://clam.iua.upf.edu/doc/CLAM-devel-doxygen/classCLAM_1_1SpectralDescriptors.html http://clam.iua.upf.edu/doc/CLAM-devel-doxygen/classCLAM_1_1EnvelopeDescriptors.html http://clam.iua.upf.edu/doc/CLAM-devel-doxygen/classCLAM_1_1SpectralPeakDescriptors.html http://clam.iua.upf.edu/doc/CLAM-devel-doxygen/classCLAM_1_1AudioTextDescriptors.html http://clam.iua.upf.edu/doc/CLAM-devel-doxygen/classCLAM_1_1SegmentDescriptors.html -- David García Garzón (Work) dgarcia at iua dot upf anotherdot es (Home) vokimon at telefonica adot net http://www.iua.upf.edu/~dgarcia
pgpwiEU0NjM69.pgp
Description: PGP signature
_______________________________________________ CLAM mailing list [email protected] http://www.iua.upf.es/mtg/clam
