Speech Enhancement Using Joint DNN-NMF Model Learned with Multi-Objective Frequency Differential Spectrum Loss Function

Figure 7

The magnitude spectrograms of noisy speech contaminated with factory noise at −5 dB SNR (a) and clean speech (b); the estimated speech by LSTM-Mask (c) and by the proposed Jnt-DNN-NMF-MOFD method (d).

(a)

(b)

(c)

(d)