摘要
目的:介绍一种新的用于非接触雷达语音增强的算法。方法:非接触检测系统自身产生的电磁噪声和电路噪声,在很大程度上降低了雷达语音信号的质量。基于提升小波良好的运算性能和时-频域分析性能,提出了采用提升小波算法来进行非接触语音增强。结果:提升算法与其他经典语音增强算法———谱减法、维纳滤波法相比,能够高效地去除雷达语音中的噪声成分,并较好地保留原始语音分量。结论:该方法可以有效地去除非空气传导语音噪声,具有运算速度快等多种优点,具有较好的应用前景。
Objective To develop a new method for enhancing the non-contact radar speech. Methods The non-contact detection system often produced electromagnetic noise and circuit noise, which reduced the quality of radar speech signal. The lifting wavelet behaved well in operation and time-frequency analysis, and thus could be used to enhance the non-contact speech,Results The lifting wavelet behaved better for eliminating the noise in the radar speech with the speech quality guaranteed, when compared with the traditional algorithms such as spectral subtraction and Wiener filter. Concluaion This method can remove the noise of non-air conduction speech and has a brilliant prospective. [Chinese Medical Equipment Journal, 2013,34 （5） ： 24-25,92]
出处
《医疗卫生装备》
CAS
2013年第5期24-25,92,共3页
Chinese Medical Equipment Journal
关键词
非接触语音
提升小波
非接触语音增强
语谱图
non-contact speech
lifting wavelet
non-contact speech enhancement
spectrogram
