Audio signal was segmented into frames and the masking thresholds of all sub-bands were estimated by psychoacoustic analysis.

• 算法将音频信号分帧进行心理声学模型分析,估计频域中各子带的最小掩蔽阈值;