通信学报 ›› 2013, Vol. 34 ›› Issue (8): 95-101.doi: 10.3969/j.issn.1000-436x.2013.08.013

• 学术论文 • 上一篇    下一篇

融合统计模型与EMD的宽带话音增强方法

周璇,鲍长春,夏丙寅   

  1. 北京工业大学 电子信息与控制工程学院 话音与音频信号处理研究室,北京 100124
  • 出版日期:2013-08-25 发布日期:2017-08-31
  • 基金资助:
    北京市教育委员会科技发展计划重点基金资助项目

Combined wideband speech enhancement method based on statistical model and EMD

Xuan ZHOU,Chang-chun BAO,Bing-yin XIA   

  1. Speech and Audio Signal Processing Lab,School of Electronic Information and Control Engineering,Beijing University of Technology,Beijing 100124,China
  • Online:2013-08-25 Published:2017-08-31
  • Supported by:
    Beijing Natural Science Foundation Program and Scientific Research Key Program of Beijing Municipal Com-mission of Education

摘要:

提出了一种融合统计模型和经验模态分解(EMD)的宽带话音增强方法。该方法首先用统计模型增强算法消除含噪话音中的主要噪声成分,然后用一种基于活动话音检测(VAD)的EMD增强算法做后处理进一步抑制残留噪声,从而使以上2种方法的优点有效地结合。在ITU-T G.160标准下对算法进行了性能测试,测试结果表明,与经典的统计模型方法相比,在不同强度的背景噪声下,增强话音的信噪比提高都较为明显。同时,在低信噪比情况下,该方法能有效抑制增强话音高频部分的音乐噪声,提高了听觉舒适度。

关键词: 宽带话音增强, 统计模型, EMD, VAD

Abstract:

A combined wideband speech enhancement method based on statistical model and empirical mode decomposition (EMD) was proposed.First,statistical model was used to eliminate the main noise component in noisy speech.Then,the residual noise was further suppressed by a post-processing module which is a speech enhancement algorithm with voice activity detection (VAD) based on EMD.The advantages of the two methods were combined effectively.The performance of the proposed method was evaluated under the standard of ITU-T G160.The experimental results indicate that the algorithm is more effective for improving the SNR in the different noise environments than classical statistical model approach.Meanwhile,in low SNR conditions,musical noise is reduced effectively,and the speech sounds more comfortable.

Key words: wideband speech enhancement, statistical model, EMD, VAD

No Suggested Reading articles found!