首页 > 《声学学报：英文版》 > 2011年3期 > A method of whispered speech enhancement based on speech absence probability and modified mel-domain masking model

A method of whispered speech enhancement based on speech absence probability and modified mel-domain masking model

在线阅读下载PDF 导出详情

摘要用听觉的掩盖的低声说的讲话改进在修改Mel-的模型领域和讲话缺席概率树液被建议。考虑到发声字符--耳语的acteristic，我们修改可伸缩的Mel频率模型。低声说的讲话被建议模型过滤。同时，为每个频率乐队掩盖阀值的价值被讲话缺席概率动态地决定。然后低声说的讲话改进被适应地用不同掩盖的阀值价值修正光谱减法系数进行。提高的低声说的信号上的客观、主观的测试的结果与另外的方法相比显示出那；建议方法能由在掩盖的阀值价值下面减少音乐噪音和背景噪音与更好主观的听觉的质量和更少的失真提高低声说的信号。

DOI 2d3o8mqnd9/988291

作者

机构地区不详

出处《声学学报：英文版》 2011年3期

关键词语音增强听觉掩蔽模型概率增强方法 MEL

分类 [理学][声学]

出版日期 2011年03月13日（中国期刊网平台首次上网日期，不代表论文的发表时间）

相关文献

1WU Hongwei;WU Zhenyang;ZHAO Li. Speech enhancement based on multitaper spectrum and psychoacoustical weighting rule.声学,2007-03.
2. ADAPTIVE MULTIRESOLUTION SPEECH ENHANCEMENT ALGORITHM BASED ON WAVELET TRANSFORM.物理电子学,1999-02.
3WANG Zhiping ZHAO Li ZOU Cairong. Speech emotion recognition based on statistical pitch model.声学,2006-01.
4LUXugang;JIAChuan;DANGJianwu. Robust speech features representation based on computational auditory model.声学,2004-04.
5. Research on hoarse speech based on vocal cord model and glottal waveform.声学,1997-01.
6Zou Xia Zhang Xiongwei. SPEECH ENHANCEMENT USING AN MMSE SHORT TIME DCT COEFFICIENTS ESTIMATOR WITH SUPERGAUSSIAN SPEECH MODELING.物理电子学,2007-03.
7HUANG Hao ZHU Jie. Tone model integration based on discriminative weight training for Putonghua speech recognition.声学,2008-03.
8YIN Hui XIE Xiang KUANG Jingming. Acoustic features based on auditory model and adaptive fractional Fourier transform for speech recognition.声学,2011-04.
9XU Shun CHEN Shaorong LIU Yulin. Blind speech source separation via nonlinear time-frequency masking.声学,2008-03.
10ZENG Qingning;OUYANG Shan. Speech enhancement by array crosstalk resistant ANC and spectrum subtraction.声学,2008-01.

来源期刊

声学学报：英文版

相关推荐

同分类资源更多

相关关键词

语音增强听觉掩蔽模型概率增强方法 MEL

返回顶部