The Priori SNR Estimation Based on the Post Frame Information and Self-Adaptive Averaging Factor
Xiaoqun Yi, Ruwei Li, Shuang Zhang
Available Online October 2016.
- https://doi.org/10.2991/ceie-16.2017.72How to use a DOI?
- A Priori SNR; ThePost Frame Speech Information; Speech Enhancement
- In view of the estimation problem of the priori signal-to-noise ratio (SNR) estimation in single-channel speech enhancement algorithm, a novel method based on the further information is proposed. In this paper, the information from the post frame is considered in obtaining priori SNR of the current frame. Firstly, the original priori SNR is achieved by the decision directed (DD) algorithm. secondly, in order to solve the smoothness drawback of Two-Step Noise Reduction (TSNR) method, we utilize the correlation of the inter-frame and introduce speech information from the post frame to refine the estimation priori SNR. Lastly, to track noisy speech change more quickly, a self-adaptive averaging factor is introduced by the minimum mean-squared error (MMSE) estimator. The proposed algorithm has good performance in reducing the speech distortion while the advantages in noise suppression are kept. The simulation experiment results show that the performance of proposed algorithm is superior than TSNR algorithm under various noise conditions.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Xiaoqun Yi AU - Ruwei Li AU - Shuang Zhang PY - 2016/10 DA - 2016/10 TI - The Priori SNR Estimation Based on the Post Frame Information and Self-Adaptive Averaging Factor BT - Proceedings of the International Conference on Communication and Electronic Information Engineering (CEIE 2016) PB - Atlantis Press SP - 561 EP - 566 SN - 2352-5401 UR - https://doi.org/10.2991/ceie-16.2017.72 DO - https://doi.org/10.2991/ceie-16.2017.72 ID - Yi2016/10 ER -