本文選題：線性預(yù)測　切入點(diǎn)：算法收斂　出處：《聲學(xué)學(xué)報(bào)》2014年05期

【摘要】：語音線性預(yù)測分析算法在噪聲環(huán)境下性能會急劇惡化,針對這一問題,提出一種改進(jìn)的噪聲魯棒稀疏線性預(yù)測算法。首先采用學(xué)生t分布對具有稀疏性的語音線性預(yù)測殘差建模,并顯式考慮加性噪聲的影響以提高模型魯棒性,從而構(gòu)建完整的概率模型。然后采用變分貝葉斯方法推導(dǎo)模型參數(shù)的近似后驗(yàn)分布,最終實(shí)現(xiàn)噪聲魯棒的稀疏線性預(yù)測參數(shù)估計(jì)。實(shí)驗(yàn)結(jié)果表明,與傳統(tǒng)算法以及近幾年提出的基于l_1范數(shù)優(yōu)化的稀疏線性預(yù)測算法相比,該算法在多項(xiàng)指標(biāo)上具有優(yōu)勢,對環(huán)境噪聲具有更好的魯棒性,并且譜失真度更小,因而能夠有效提高噪聲環(huán)境下的語音質(zhì)量。
[Abstract]:The performance of speech linear predictive analysis algorithm will deteriorate rapidly in noise environment. To solve this problem, an improved robust sparse linear prediction algorithm is proposed.First, the student t distribution is used to model the sparse linear prediction residual of speech, and the effect of additive noise is explicitly considered to improve the robustness of the model and to construct a complete probabilistic model.Then the variational Bayesian method is used to derive the approximate posterior distribution of the model parameters, and finally the robust sparse linear prediction parameter estimation of noise is realized.The experimental results show that compared with the traditional algorithm and the sparse linear prediction algorithm based on l-1 norm optimization proposed in recent years, the algorithm has many advantages, such as better robustness to environmental noise, and less spectral distortion.Therefore, the speech quality in noisy environment can be improved effectively.
【作者單位】： 解放軍理工大學(xué);
【基金】：江蘇省自然科學(xué)基金(BK2012510) 國家博士后科研基金(20090461424)資助
【分類號】：TN912.3

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 陳淑珍,張晨光,劉懷林,張q，

本文編號：1713916