Review Article
A New Speech Enhancement Technique Based on Stationary Bionic Wavelet Transform and MMSE Estimate of Spectral Amplitude
Table 8
Results obtained from the computation of SNR (signal 3 (male voice) corrupted by tank noise).
| | | The speech enhancement technique | Unsupervised speech denoising via perceptually motivated robust principal component analysis [23] | The speech enhancement technique based on MSS-SMPO [24, 25] | The proposed speech enhancement technique | The denoising technique based on MMSE Estimate of Spectral Amplitude [22] |
| ā5 | 1.8657 | 2.8506 | 4.5261 | 3.0084 | 0 | 5.2569 | 6.0328 | 8.2533 | 6.6513 | 5 | 8.8912 | 9.7367 | 12.5241 | 10.7510 | 10 | 12.5670 | 13.6433 | 16.6318 | 14.7634 | 15 | 15.9158 | 18.2296 | 21.0643 | 19.0401 |
|
|
The bold values show the values obtained from the application of the proposed speech enhancement technique, and they are the best values.
|