Audio Fingerprinting MUMT 611 Ichiro Fujinaga Mc Gill

  • Slides: 11
Download presentation
Audio Fingerprinting MUMT 611 Ichiro Fujinaga Mc. Gill University

Audio Fingerprinting MUMT 611 Ichiro Fujinaga Mc. Gill University

Introduction § Fingerprints uniquely identify people § Audio fingerprints aims to uniquely identify a

Introduction § Fingerprints uniquely identify people § Audio fingerprints aims to uniquely identify a piece of music from a short excerpt of the music § Other names: v. Acoustic fingerprinting v. Content-based audio identification MUMT 611 Fujinaga 2 / 11

Applications § “The popular social networking site My. Space. com announced Monday that it

Applications § “The popular social networking site My. Space. com announced Monday that it has licensed technology [Gracenote] that will help it prevent unauthorized copyrighted music from being posted to My. Space users’ pages. ” Macworld (2006/10/06) § “Adding missing album art: With the increased emphasis on album art, Windows Media Player 11 also ensures that missing album art isn't a problem. Most album art can automatically be populated in the background using the advanced audio fingerprinting capabilities in Windows Media Player 11. ” www. microsoft. com/windowsmedia/player/11 MUMT 611 Fujinaga 3 / 11

MUMT 611 Fujinaga 4 / 11

MUMT 611 Fujinaga 4 / 11

Commercial products § Gracenote § M 2 any § Audible Magic (Muscle Fish) MUMT

Commercial products § Gracenote § M 2 any § Audible Magic (Muscle Fish) MUMT 611 Fujinaga 5 / 11

Basic framework (Cano et al. 2005) MUMT 611 Fujinaga 6 / 11

Basic framework (Cano et al. 2005) MUMT 611 Fujinaga 6 / 11

Challenges § Variance v. Compression v. Distortion v. Noise GOALS §Robust § Efficiency §Compact

Challenges § Variance v. Compression v. Distortion v. Noise GOALS §Robust § Efficiency §Compact v. Encoding v. Loopkup §Fast ØDatabase size ØSearch algorithm § Music High dimensionality MUMT 611 Fujinaga 7 / 11

Extraction Fingerprint extraction (Cano et al. 2005) MUMT 611 Fujinaga 8 / 11

Extraction Fingerprint extraction (Cano et al. 2005) MUMT 611 Fujinaga 8 / 11

Searching § § § Euclidean / HMM sequence Pre-computed distances Multi-staged searching (coarse to

Searching § § § Euclidean / HMM sequence Pre-computed distances Multi-staged searching (coarse to fine) Indexing Candidate pruning Table lookup MUMT 611 Fujinaga 9 / 11

Table lookup database (Haitsma et al. 2002) MUMT 611 Fujinaga 10 / 11

Table lookup database (Haitsma et al. 2002) MUMT 611 Fujinaga 10 / 11

References § Cano, P. , E. Batlle, T. Kalker, and J. Haitsma. 2005. A

References § Cano, P. , E. Batlle, T. Kalker, and J. Haitsma. 2005. A review of audio fingerprinting. Journal of VLSI Signal Processing Systems 41 (3): 271– 84. § Haitsma, J. , and T. Kalker. 2002. A highly robust audio fingerprinting system. Proceedings of the International Conference on Music Information Retrieval. 107– 15. MUMT 611 Fujinaga 11 / 11