基于隐马尔可夫模型的非监督噪声功率谱估计

Unsupervised Noise Power Estimation Using Hidden Markov Model

噪声功率谱估计是语音增强算法的基本组成部分,传统算法大多采用启发式的估计方法,因而不能保证噪声估计值的统计最优。提出了一种基于极大似然的非监督噪声功率谱估计方法,采用隐马尔可夫模型(Hidden Markov model,HMM)在每个子带建立语音和非语音对数功率谱的统计模型,模型包含语音和非语音两个高斯分量,其中非语音高斯分量的均值表示噪声功率谱估计值,根据最大期望(Expectation maximization,EM)算法得到包括噪声均值在内的HMM参数集。针对语音信号可能出现的长时缺失,对HMM引入了一些约束条件,保证了模型的稳定性。实验表明,该方法获得的极大似然噪声估计优于基于启发式的经典方法获得的噪声估计。

Noise estimation is a fundamental part of speech enhancement.Most traditional methods are heuristic which can not enable the optimal estimation.An unsupervised noise power estimation is presented based on maximum likelihood.A log-power statistical model is constructed using hidden Markov model（HMM）in each subband.This model comprises speech and nonspeech Gauss components,and the mean value of nonspeech Gauss component is the estimation of noise power.Moreover,speech may be long-term absent,some constraints are introduced to this model for stability.The experiments validate that the proposed method can obtain the maximum likelihood noise estimation and outperforms conventional heuristic methods.