[1] Wu Hongwei, Wu Zhenyang, Zhao Li, et al. Improved mean-shift-based pitch determination [J]. Journal of Southeast University (English Edition), 2007, 23 (4): 494-499. [doi:10.3969/j.issn.1003-7985.2007.04.004]
Copy
Improved mean-shift-based pitch determination(
)
改进的基于均值移动的基音检测
)Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]
- Volumn:
- 23
- Issue:
- 2007 4
- Page:
- 494-499
- Research Field:
- Information and Communication Engineering
- Publishing date:
- 2007-12-30
Info
- Title:
- Improved mean-shift-based pitch determination
- 改进的基于均值移动的基音检测
- Author(s):
- Wu Hongwei1; 2; Wu Zhenyang1; Zhao Li1
-
1 School of Information Science and Engineering, Southeast University, Nanjing 210096, China
2 School of Electronics and Information, Suzhou University, Suzhou 215021, China
- Keywords:
- pitch; pitch determination; mean shift algorithm
- PACS:
- TN912.3
- DOI:
- 10.3969/j.issn.1003-7985.2007.04.004
- Abstract:
- The underlying principle of pitch determination based on the mean shift algorithm is studied, and the cause of pitch error propagation in the original pseudo code is analyzed.The problem of error propagation is solved by choosing an appropriate initial pitch candidate F00.The theoretical choice guideline in a pitch epoch is obtained as ensuring the true pitch F0 satisfying F00/002<F0<03F00/002.The validity of the choice guideline is verified by the F0000 experiment.Meanwhile, the algorithm is extended to the pitch determination in the noisy case and compared with the method of subharmonic-to-harmonic ratio(SHR).The experimental results show that the improved algorithm bears comparison with SHR and it runs much faster than SHR.
- 研究了使用均值移动算法进行基音检测的基本原理, 分析了原始伪码中基音错误传播的原因, 通过选择一合适的基音初始值F00解决了这一问题.理论上推导了在一有声段内基音初始值的选取原则, 即使实际基音F00满足F00/002<F0<03F00/002.然后通过实验验证了初始基音选取原则的正确性.同时将这一算法推广到噪声情形下的基音检测, 并将其与子谐波谐波比(subharmonic-to-harmonic ratio, SHR)方法进行了对比, 各种信噪比下的实验结果表明该方法与SHR方法可比而且运行速度更快.
References:
[1] Zwicker E, Fastl H.Psychoacoustics:facts and models[M].2nd ed.Berlin:Springer-Verlag, 1999:118-122.
[2] Bagshaw P C, Hiller S M, Jack M A.Enhanced pitch tracking and the processing of F0 contours for computer aided intonation teaching [C]//Proc 3rd European Conf on Speech Communication and Technology.Berlin, Germany, 1993:1003-1006.
[3] Veprek P, Scordilis M S.Analysis, enhancement and evaluation of five pitch determination techniques [J].Speech Communication, 2002, 37(3):249-270.
[4] Sun X.Pitch determination and voice quality analysis using subharmonic-to-harmonic ratio [C]//ICASSP 2002.Orlando, Florida, USA, 2002:333-336.
[5] Hasan M K, Hussain S, Setu M T H, et al.Signal reshaping using dominant harmonic for pitch estimation of noisy speech [J].Signal Processing, 2006, 86(5):1010-1018.
[6] Rouat J, Liu Y C, Morissette D.A pitch determination and voiced/unvoiced decision algorithm for noisy speech [J].Speech Communication, 1997, 21(3):191-207.
[7] Képesi M, Weruaga L.High-resolution noise-robust spectral-based pitch estimation [C]//Eurospeech 2005.Lisboa, Portugal, 2005:313-316.
[8] Luo Yafei, Bao Changchun.Super resolution pitch detection based on band-partitioning spectral entropy and signal decomposition in DCT domain [J]. Acta Electronica Sinica, 2007, 35(1):13-22.(in Chinese)
[9] Weruaga L, Képesi M.Speech analysis with the fast chirp transform[C]//European Signal Processing Conf (EUSIPCO).Vienna, Austria, 2004:1011-1014.
[10] Fukunaga K, Hostetler L D.The estimation of the gradient of a density function, with applications in pattern recognition [J].IEEE Trans Information Theory, 1975, 21(1):32-40.
[11] Cheng Y.Mean shift, mode seeking, and clustering [J].IEEE Trans Pattern Analysis and Machine Intelligence, 1995, 17(8):790-799.
[12] Comaniciu D, Meer P.Mean shift:a robust approach toward feature space analysis [J].IEEE Trans Pattern Analysis and Machine Intelligence, 2002, 24(5):603-619.
Memo
- Memo:
- Biographies: Wu Hongwei(1967—), female, graduate;Wu Zhenyang(corresponding author), male, professor, zhenyang@seu.edu.cn.