저작권 보호를 위한 HMM기반의 음악 식별 시스템
HMM-based Music Identification System for Copyright Protection
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
In this paper, in order to protect music copyrights, we propose a music identification system which is scalable to the number of pieces of registered music and robust to signal-level variations of registered music. For its implementation, we define the new concepts of ‘music word’ and ‘music phoneme’ as recognition units to construct ‘music acoustic models’. Then, with these concepts, we apply the HMM-based framework used in continuous speech recognition to identify the music. Each music file is transformed to a sequence of 39-dimensional vectors. This sequence of vectors is represented as ordered states with Gaussian mixtures. These ordered states are trained using Baum-Welch re-estimation method. Music files with a suspicious copyright are also transformed to a sequence of vectors. Then, the most probable music file is identified using Viterbi algorithm through the music identification network. We implemented a music identification system for 1,000 MP3 music files and tested this system with variations in terms of MP3 bit rate and music speed rate. Our proposed music identification system demonstrates robust performance to signal variations. In addition, scalability of this system is independent of the number of registered music files, since our system is based on HMM method.
Abstract
In this paper, in order to protect music copyrights, we propose a music identification system which is scalable to the number of pieces of registered music and robust to signal-level variations of registered music. For its implementation, we define the new concepts of ‘music word’ and ‘music phoneme’ as recognition units to construct ‘music acoustic models’. Then, with these concepts, we apply the HMM-based framework used in continuous speech recognition to identify the music. Each music file is transformed to a sequence of 39-dimensional vectors. This sequence of vectors is represented as ordered states with Gaussian mixtures. These ordered states are trained using Baum-Welch re-estimation method. Music files with a suspicious copyright are also transformed to a sequence of vectors. Then, the most probable music file is identified using Viterbi algorithm through the music identification network. We implemented a music identification system for 1,000 MP3 music files and tested this system with variations in terms of MP3 bit rate and music speed rate. Our proposed music identification system demonstrates robust performance to signal variations. In addition, scalability of this system is independent of the number of registered music files, since our system is based on HMM method.
참고문헌 (8)
- J. Haitsma, "A highly robust audio fingerprinting system", International Conference on Music Information Retrieval (ISMIR), pp.144~148 (2002)
- M. Miller, "Audio fingerprinting: nearest neighbor search in high dimensional binary spaces", IEEE International Workshop on Multimedia Signal Processing, pp.182~185 (2002)
- R. Typke, "A survey of music information retrieval systems", International Conference on Music Information Retrieval, pp.153~160 (2005)
- W. Glaser, "Consumer item matching method and system", U.S. Patent No. 7003515, ~
- A. Ghias, "Query by humming - musical information retrieval in an audio database", ACM Multimedia, pp.231~236 (1995)
- E. Unal, "A statistical approach to retrieval under user-dependent uncertainty in query-by-humming systems", ACM SIGMM International Workshop on Multimedia Information Retrieval, pp.113~118 (2004)
- S.Young,G.Evermann,M.Gales et al., "The HTK book(for HTK version 3.4)", Cambridge University., ~ (2006)