Using Paired Distances of Signal Peaks in Stereo Channels as Fingerprints for Copy Identification

Abstract

This article proposes to use the relative distances between adjacent envelope peaks detected in stereo audio as fingerprints for copy identification. The matching algorithm used is the rough longest common subsequence (RLCS) algorithm. The experimental results show that the proposed approach has better identification accuracy than an MPEG-7 based scheme for distorted and noisy audio. When compared with other schemes, the proposed scheme uses fewer bits with comparable performance. The proposed fingerprints can also be used in conjunction with the MPEG-7 based scheme for lower computational burden.

Comparative Study of Dimensionality Reduction Techniques for Spectral–Temporal Data

“…Music identification [2,3,15] is a technique to identify an audio clip from a large set of soundtracks based on the signal waveform. Therefore, even if the same person sings the same songs two times, these two audio works are considered as different.…”

Simulated smart phone recordings for audio identification

“…There are many existing schemes to compute fingerprints [9,[12][13][14][15][16], some of which are briefly described in the following. Haitsma and Kalker proposed a fingerprinting scheme based on binarized spectral values in each audio frame [12].…”

“…A fingerprinting scheme, proposed by Ramona and Peeters, used cosine filters to deal with the problem of pitch shifting and used the energy of the spectrum to perform frame synchronization [15]. For stereo music, You and Pu proposed a scheme to compute the fingerprints solely based on the distances between two successive waveform peaks [16]. Other than the mentioned approaches, there are more than ten proprietary fingerprinting schemes [17] currently available.…”