Abstract
This paper studies the multichannel maximum signal-to-noise ratio (SNR) filter for noise reduction in the short-time Fourier transform (STFT) domain. By considering both the interchannel and interframe information, the maximum SNR filter is formulated by maximizing the output SNR with the minimum distortion constraint. Simulations are performed to examine the performance of the deduced multichannel maximum SNR noise reduction filter in different conditions and the results demonstrated that this filter can significantly increase both the SNR and the speech quality.
| Originalsprog | Engelsk |
|---|---|
| Titel | 2015 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2015 - Proceedings |
| Antal sider | 4 |
| Forlag | IEEE |
| Publikationsdato | 31 aug. 2015 |
| Sider | 697-700 |
| Artikelnummer | 7230494 |
| ISBN (Trykt) | 9781479919482 |
| DOI | |
| Status | Udgivet - 31 aug. 2015 |
| Udgivet eksternt | Ja |
| Begivenhed | IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2015 - Chengdu, Kina Varighed: 12 jul. 2015 → 15 jul. 2015 |
Konference
| Konference | IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2015 |
|---|---|
| Land/Område | Kina |
| By | Chengdu |
| Periode | 12/07/2015 → 15/07/2015 |
| Sponsor | Institute of Electrical and Electronics Engineers Signal Processing Society (SPS) |