A binaural speech enhancement algorithm: Application to background and directional noise fields

In this paper, we presented a binaural speech enhancement algorithm that is effective in eliminating both directional and background noises. Firstly, we utilized a speech presence probability function to estimate the PSD (Power Spectral Density) of target speech and background noises. The function is based on power-level-differences. In this way, a coherence-based dual-microphone algorithm was used in binaural background noises reduction. Then we suppressed the directional interference noises by using the defined speech presence probability function via the time-frequency (T-F) masking process. The simulation results show that the method is effective.