Research Article
[Retracted] Research and DSP Implementation of Speech Enhancement Technology Based on Dynamic Mixed Features and Adaptive Mask
Table 2
Average Short-Term Objective Intelligibility of Speech (STOI).
| Noise | Signal-to-noise ratio (dB) | SRNN | SRU | GRU | CGRU |
| factory2 | −5 | 0.773 | 0.789 | 0.786 | 0.805 | 0 | 0.852 | 0.868 | 0.870 | 0.888 | 5 | 0.901 | 0.916 | 0.912 | 0.929 | buccaneer1 | −5 | 0.623 | 0.625 | 0.584 | 0.636 | 0 | 0.745 | 0.748 | 0.720 | 0.760 | 5 | 0.833 | 0.836 | 0.826 | 0.852 | destroyerengine | −5 | 0.624 | 0.616 | 0.595 | 0.633 | 0 | 0.749 | 0.752 | 0.734 | 0.767 | 5 | 0.841 | 0.850 | 0.841 | 0.865 | hfchannel | −5 | 0.655 | 0.655 | 0.645 | 0.671 | 0 | 0.768 | 0.774 | 0.773 | 0.792 | 5 | 0.847 | 0.862 | 0.861 | 0.879 |
|
|