【摘要】：針對(duì)現(xiàn)有手機(jī)雙麥克風(fēng)消噪系統(tǒng)無(wú)法應(yīng)對(duì)多種復(fù)雜噪聲環(huán)境、在消除噪聲的同時(shí)會(huì)引起語(yǔ)音失真等問(wèn)題,本文提出一種新的手機(jī)雙麥克風(fēng)消噪系統(tǒng)。該系統(tǒng)將時(shí)域與頻域處理結(jié)合,在噪聲估計(jì)和消除兩方面均有改進(jìn),結(jié)合單、雙麥克風(fēng)的噪聲估計(jì)算法,提高了噪聲估計(jì)的準(zhǔn)確性。此外,將基音檢測(cè)與消噪處理結(jié)合,在語(yǔ)音幀中估計(jì)語(yǔ)音基音頻率,確定語(yǔ)音和噪聲頻率點(diǎn),根據(jù)語(yǔ)音和噪聲頻率點(diǎn)分別調(diào)整維納濾波器參數(shù),在濾除噪聲的同時(shí)盡可能地保留語(yǔ)音頻率點(diǎn),從而減少語(yǔ)音失真。實(shí)驗(yàn)結(jié)果表明,與現(xiàn)有雙麥克風(fēng)消噪系統(tǒng)相比,本系統(tǒng)在抑制噪聲的同時(shí)能夠有效減少消噪算法對(duì)語(yǔ)音造成的損害,提高了手機(jī)通話質(zhì)量,對(duì)于方向性的語(yǔ)音干擾也起到很好的抑制效果。
[Abstract]:In order to solve the problems that the existing mobile phone dual microphone denoising system can not cope with a variety of complex noise environments and will cause speech distortion while eliminating noise, a new mobile phone dual microphone denoising system is proposed in this paper. The system combines time domain and frequency domain processing, and improves the accuracy of noise estimation by combining the noise estimation algorithm of single microphone and double microphone. In addition, the pitch detection and denoising processing are combined to estimate the speech pitch frequency in the speech frame, determine the speech and noise frequency points, and adjust the Wiener filter parameters according to the speech and noise frequency points, respectively. The speech frequency points are preserved as much as possible while filtering the noise, so as to reduce the speech distortion. The experimental results show that compared with the existing dual microphone denoising system, this system can effectively reduce the voice damage caused by denoising algorithm and improve the quality of mobile phone calls. It also has a good suppression effect on directional speech interference.
【作者單位】： 南京師范大學(xué)物理與科學(xué)技術(shù)學(xué)院江蘇省音頻技術(shù)工程實(shí)驗(yàn)室;
【基金】：新一代智能手機(jī)多麥克風(fēng)消噪音頻測(cè)試與輔助設(shè)計(jì)系統(tǒng)關(guān)鍵技術(shù)研究(BE2014139)
【分類號(hào)】：TN912.3

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 夏丙寅;鮑長(zhǎng)春;;適應(yīng)噪聲強(qiáng)度突變的噪聲估計(jì)加速方法[J];信號(hào)處理;2013年10期

2 朱磊;徐佩霞;何佳;;一種高效穩(wěn)健的差分視頻噪聲估計(jì)算法[J];數(shù)據(jù)采集與處理;2006年03期

3 劉斌;戎蒙恬;劉文江;王若琳;;一種快速且準(zhǔn)確的噪聲估計(jì)算法[J];信息技術(shù);2012年01期

4 鄒霞;趙繼勇;張雄偉;;一種基于χ~2分布的子帶噪聲估計(jì)方法[J];數(shù)據(jù)采集與處理;2008年01期

5 李光源;崔慧娟;唐昆;;一種基于噪聲估計(jì)的語(yǔ)音激活檢測(cè)算法[J];信息技術(shù);2011年10期

6 黃騰飛;李幫義;熊季霞;;基于小波方差分解的混沌時(shí)間序列噪聲估計(jì)和閾值去噪[J];計(jì)算機(jī)應(yīng)用;2013年03期

7 賀慧芳;馬建芬;;基于語(yǔ)音幀檢測(cè)和子帶譜跟蹤的噪聲估計(jì)算法[J];計(jì)算機(jī)工程;2009年20期

8 吳鵬;李晶皎;王愛(ài)俠;王驕;;相位匹配噪聲估計(jì)的高階譜去噪方法[J];小型微型計(jì)算機(jī)系統(tǒng);2010年12期

9 徐立中;馮肖亮;文成林;;網(wǎng)絡(luò)化多傳感器系統(tǒng)的序貫式融合濾波——基于噪聲估計(jì)方法[J];電子學(xué)報(bào);2014年01期

10 陳會(huì)娟;戴聲奎;;改進(jìn)的基于分塊法的圖像噪聲估計(jì)[J];計(jì)算機(jī)應(yīng)用;2014年07期

相關(guān)會(huì)議論文 前5條

1 鄒霞;陳亮;張雄偉;;一種改進(jìn)的最小統(tǒng)計(jì)量控制遞歸平均噪聲估計(jì)算法[A];第十二屆全國(guó)信號(hào)處理學(xué)術(shù)年會(huì)（CCSP-2005）論文集[C];2005年

2 崔杰;張?jiān)埔?王s，

本文編號(hào)：2488757