基于非均匀谱压缩特征的模型补偿新算法
doi: 10.3724/SP.J.1146.2005.01316
Novel Model Compensation Based on Non-uniform Spectral Compression Features
-
摘要: 在信噪比依赖的非均匀谱压缩(SNSC)鲁棒语音特征提取技术和VTS算法的基础上，该文提出了一种新的MC-SNSC模型补偿算法。SNSC技术是一种根据人类听觉对声音强度-响度感知转化关系的谱幅度变化操作和噪声抑制技术。基于对数谱域的噪声以及SNSC特征提取对语音信号特征所产生的失配函数，推导出了MC-SNSC模型补偿算法。实验证明使用这一新算法，识别率比当前较理想的VTS和PMC算法有很明显的提升，算法的复杂度较VTS等算法仅有轻微的增加。
-
关键词:
- 语音识别;模型补偿;非均匀谱压缩
Abstract: A novel model compensation method is proposed, which integrates the Vector Taylor Series (VTS) approach with a robust feature extraction technique called SNR-dependent Non-uniform Spectral Compression (SNSC). The SNSC method is a spectral operation of magnitude transformation which resembles the human intensity-to-loudness conversion process and de-emphasizes noisy bands. Based on this mismatch function, which models the effect of the noise onto the clean speech in the Log-spectral domain together with the SNSC, a new model compensation procedure is derived. By adopting this novel model compensation approach, significant improvement over the PMC and VTS method can be found in different additive noisy environments at the expense of slight increase in computational complexity. -
罗宇, 杜利民. 基于单高斯模型集的汉语美子带特征重建算法. 电子学报, 2004, 32(10): 1654-1657. Luo Yu and Du Li-min. Single Gauss model set based MAP data imputation method for Mel-frequency filter-bank vectors of chinese speech. Acta Electronica Sinica, 2004, 32(10): 1654-1657.[2]Ding Pei and Cao Z G. An efficient robust ASR system based on the combination of speech enhancement and HMM adaptation. Chinese Journal of Electronics, 2002, 11(3): 422-425.[3]Acero A, Deng L, Kristjansson T, and Zhang J. HMM adaptation using vector Taylor series for noise speech recognition. in Proc. ICSLP2000, Beijing, China, Oct. 2000: 869-872.[4]Hung J W, Shen J L, and Lee L S. New approach for domain transformation and parameter combination for improved accuracy in parallel model combination (PMC) techniques[J].IEEE Trans. on Speech and Audio Processing.2001, 9(8):842-854[5]Gong Y. Speech recognition in noisy environments: Asurvey[J].Speech Communication.1995, 16(3):261-291[6]Chu K K and Leung S H. SNR-dependent non-uniform spectral compression for noisy speech recognition. In Proc. ICASSP04, Montreal, Canada, May 2004: 973-976.[7]Abramowitz M and Stegun I A. Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables. New York: Dover Publications Inc., 1972. -
计量
- 文章访问数: 3572
- HTML全文浏览量: 101
- PDF下载量: 567
- 被引次数: 0
下载:
