MASK+: Data-driven regions selection for acoustic fingerprinting

Audio fingerprinting, content recognition

In this paper we propose an improvement to MASK, a recently proposed acoustic fingerprint that has been shown to be effective at compactly representing an acoustic signal using binary descriptors.

Acoustic fingerprinting is the process to deterministically obtain a compact representation of an audio segment, used to compare multiple audio files or to efficiently search for a file within a big database. Recently, we proposed a novel fingerprint named MASK (Masked Audio Spectral Keypoints) that encodes the relationship between pairs of spectral regions around a single spectral energy peak into a binary representation. In the original proposal the configuration of location and size of the regions pairs was determined manually to optimally encode how energy flows around the spectral peak. Such manual selection has always been considered as a weakness in the process as it might not be adapted to the actual data being represented. In this paper we address this problem by proposing a unsupervised, data-driven method based on mutual information theory to automatically define an optimal MASK fingerprint structure. Audio retrieval experiments optimizing for data distorted with additive Gaussian white noise show that the proposed method is much more robust than the original MASK and a well known acoustic fingerprint