A Highly Robust Audio Monitoring System for Radio Broadcasting

Abstract

Proposing a novel approach for monitoring
songs for the radio broadcasting channels is very
important for the interest of singers, writers and
musicians in the musical industry. Singers, writers and
musicians have a claim to intellectual property rights
for their songs broadcast over all the radio channels.
According to this intellectual property rights act
singers, writers and musicians should be paid for their
songs broadcast over all the radio channels. Therefore we
propose a real time audio monitoring approach to solve
this problem which includes our own audio recognition
algorithm. It is easy to recognize a song, when you provide
the original high quality blueprint of the song as input. But
we can’t expect such kind of audio input from radio
channels since lots of transformations are possible before
reaching the end user or listener. For example, adding
environmental effects such as noise, adding commercials
on the song as watermarks, playing more than one song
as a chain without adding any silence between them,
playing a part of the song, playing same song in various
speeds and so on. These transformations cause change in
the uniqueness of particular song and make the problem
even more difficult. The algorithm we proposing is resistant
to noise and distortion as well as it is capable of recognizing
short segment of song when broadcasting over the radio
channels. At the end of the processing our system generates
a descriptive report including title of the song, singer of the
song, writer of the song, composer of the song, number of
times it was played and when it was played for all songs for
a particular period for all radio broadcasting channels. We
evaluate our system against various types of real time
scenarios and achieved overall higher level of accuracy
(96%) at the end.