The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Este artigo propõe um algoritmo que estima adaptativamente a variação do ruído variável no tempo usado na filtragem de Kalman para aprimoramento do sinal de fala em tempo real. No sinal de voz contaminado por ruído branco, espera-se que os componentes espectrais, exceto os dominantes na banda de alta frequência, reflitam a energia do ruído. Nossa abordagem é primeiro encontrar as bandas de energia dominantes no espectro da fala usando LPC. Calculamos então o valor médio dos componentes espectrais reais na região de alta frequência, excluindo as bandas de energia dominantes, e o usamos como variação de ruído. A estimativa de variância de ruído resultante é então aplicada à filtragem de Kalman para suprimir o ruído de fundo. Os resultados experimentais indicam que a abordagem proposta alcança uma melhoria significativa em termos de aprimoramento da fala em relação à filtragem de Kalman convencional que utiliza a potência média do ruído apenas no intervalo de silêncio. Como um refinamento de nossos resultados, empregamos filtragem múltipla de Kalman com múltiplos modelos de ruído e melhoramos a inteligibilidade.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Wooil KIM, Hanseok KO, "Noise Variance Estimation for Kalman Filtering of Noisy Speech" in IEICE TRANSACTIONS on Information,
vol. E84-D, no. 1, pp. 155-160, January 2001, doi: .
Abstract: This paper proposes an algorithm that adaptively estimates time-varying noise variance used in Kalman filtering for real-time speech signal enhancement. In the speech signal contaminated by white noise, the spectral components except dominant ones in high frequency band are expected to reflect the noise energy. Our approach is first to find the dominant energy bands over speech spectrum using LPC. We then calculate the average value of the actual spectral components over the high frequency region excluding the dominant energy bands and use it as the noise variance. The resulting noise variance estimate is then applied to Kalman filtering to suppress the background noise. Experimental results indicate that the proposed approach achieves a significant improvement in terms of speech enhancement over those of the conventional Kalman filtering that uses the average noise power over silence interval only. As a refinement of our results, we employ multiple-Kalman filtering with multiple noise models and improve the intelligibility.
URL: https://global.ieice.org/en_transactions/information/10.1587/e84-d_1_155/_p
Copiar
@ARTICLE{e84-d_1_155,
author={Wooil KIM, Hanseok KO, },
journal={IEICE TRANSACTIONS on Information},
title={Noise Variance Estimation for Kalman Filtering of Noisy Speech},
year={2001},
volume={E84-D},
number={1},
pages={155-160},
abstract={This paper proposes an algorithm that adaptively estimates time-varying noise variance used in Kalman filtering for real-time speech signal enhancement. In the speech signal contaminated by white noise, the spectral components except dominant ones in high frequency band are expected to reflect the noise energy. Our approach is first to find the dominant energy bands over speech spectrum using LPC. We then calculate the average value of the actual spectral components over the high frequency region excluding the dominant energy bands and use it as the noise variance. The resulting noise variance estimate is then applied to Kalman filtering to suppress the background noise. Experimental results indicate that the proposed approach achieves a significant improvement in terms of speech enhancement over those of the conventional Kalman filtering that uses the average noise power over silence interval only. As a refinement of our results, we employ multiple-Kalman filtering with multiple noise models and improve the intelligibility.},
keywords={},
doi={},
ISSN={},
month={January},}
Copiar
TY - JOUR
TI - Noise Variance Estimation for Kalman Filtering of Noisy Speech
T2 - IEICE TRANSACTIONS on Information
SP - 155
EP - 160
AU - Wooil KIM
AU - Hanseok KO
PY - 2001
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E84-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2001
AB - This paper proposes an algorithm that adaptively estimates time-varying noise variance used in Kalman filtering for real-time speech signal enhancement. In the speech signal contaminated by white noise, the spectral components except dominant ones in high frequency band are expected to reflect the noise energy. Our approach is first to find the dominant energy bands over speech spectrum using LPC. We then calculate the average value of the actual spectral components over the high frequency region excluding the dominant energy bands and use it as the noise variance. The resulting noise variance estimate is then applied to Kalman filtering to suppress the background noise. Experimental results indicate that the proposed approach achieves a significant improvement in terms of speech enhancement over those of the conventional Kalman filtering that uses the average noise power over silence interval only. As a refinement of our results, we employ multiple-Kalman filtering with multiple noise models and improve the intelligibility.
ER -