| 采用性别相关的深度神经网络及非负矩阵分解模型用于单通道语音增强 |
| |
| 引用本文： | 李煦,王子腾,王晓飞,付强,颜永红.采用性别相关的深度神经网络及非负矩阵分解模型用于单通道语音增强[J].声学学报,2019,44(2):221-230. |
| |
| 作者姓名： | 李煦 王子腾 王晓飞 付强 颜永红 |
| |
| 作者单位： | 1. 中国科学院声学研究所 语言声学与内容理解重点实验室 北京 100190; |
| |
| 基金项目： | 国家自然科学基金项目(11461141004,61271426,U1536117,11504406,11590770-4)中国科学院战略性先导科技专项项目(XDA06030100,XDA06030500,XDA06040603)国家973计划项目(2013CB329302)新疆维吾尔自治区科技重大专项项目(201230118-3)资助国家863计划项目(2015AA016306) |
| |
| 摘 要： | 为了从带噪信号中得到纯净的语音信号,提出了一种采用性别相关模型的单通道语音增强算法。具体而言,在训练阶段,分别训练了与性别相关的深度神经网络-非负矩阵分解模型用于估计非负矩阵分解中的权重参数;在测试阶段,提出了一种基于非负矩阵分解和组稀疏惩罚的算法用于判断测试语音中说话人的性别信息,然后再采用对应的模型估计权重,并结合已训练好的字典进行语音增强。实验结果表明所提算法在噪声抑制量及语音质量上,均优于一些基于非负矩阵分解的算法和基于深度神经网络的算法。
|
| 关 键 词： | 语音增强 非负矩阵分解 深度神经网络 性别信息 |
| 收稿时间： | 2017-03-26 |
Single-channel speech enhancement based on gender-related deep neural networks and non-negative matrix factorization models |
| |
| Institution: | 1. Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190;2. University of Chinese Academy of Sciences, Beijing 100190;3. Xinjiang Laboratory of Minority Speech and Language Information Processing, Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 830011 |
| |
| Abstract: | In order to obtain the clean speech from the noisy signal, a single-channel speech enhancement algorithm based on gender-related models is proposed. Specifically, in the training stage, Deep Neural Networks(DNN) and Nonnegative Matrix Factorization(NMF) are employed to train two gender-related DNN-NMF models using the genderspecific training data. In the test stage, an algorithm based on NMF and group sparsity penalty is proposed to identify the gender information of the speaker in the test signal. Then the corresponding DNN-NMF model is used to estimate the activations for speech enhancement. Experimental results show that the proposed algorithm performs better in suppressing the noises without decreasing the speech quality compared with other NMF-based and DNN-based methods. |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《声学学报》浏览原始摘要信息 |
| 点击此处可从《声学学报》下载免费的PDF全文 |
|
