Research Article
[Retracted] Research and DSP Implementation of Speech Enhancement Technology Based on Dynamic Mixed Features and Adaptive Mask
Table 1
Average Perceptual Evaluation of Speech Quality (PESQ).
| Noise | Signal-to-noise ratio (dB) | SRNN | SRU | GRU | CGRU |
| factory2 | −5 | 2.346 | 2.515 | 2.426 | 2.495 | 0 | 2.651 | 2.806 | 2.729 | 2.843 | 5 | 2.877 | 3.102 | 3.065 | 3.127 | buccaneer1 | −5 | 1.861 | 1.952 | 1.763 | 1.952 | 0 | 2.240 | 2.333 | 2.201 | 2.347 | 5 | 2.533 | 2.654 | 2.587 | 2.668 | destroyerengine | −5 | 1.903 | 2.030 | 2.106 | 2.228 | 0 | 2.212 | 2.296 | 2.240 | 2.329 | 5 | 2.521 | 2.578 | 2.526 | 2.600 | hfchannel | −5 | 1.801 | 1.952 | 1.706 | 1.835 | 0 | 2.157 | 2.177 | 2.111 | 2.285 | 5 | 2.469 | 2.513 | 2.502 | 2.599 |
|
|