节点文献
语音模糊增强研究
Research on Fuzzy Enhancement of Speech Signal
【摘要】 针对加性有色噪声,提出了语音信号模糊增强算法;建立并训练了一个自适应神经模糊推理系统(ANFIS)的语音模糊增强系统;用其对含噪语音中的有色噪声进行了模糊估计,从而提取了干净的语音。对算法进行了仿真实验,结果表明,对模拟有色噪声在负17 dB时能提取出清晰的语音。
【Abstract】 Aiming at additive colored noise,a new fuzzy enhancement algorithm of speech signal based on the principle of fuzzy inference in fuzzy mathematics was presented.A fuzzy enhancement system of speech signal——Auto Neural Fuzzy Inference System(ANFIS) was set up,which can be used to make fuzzy estimation to the colored noise contained in noises to extract clean speech signal.The results of the simulated experiment on the algorithm for different signals to noise ratio show that the system can extract clean speech signal from the simulated colored noise at-17 dB of signal to noise ratio.
【Key words】 speech signal; colored noise; white noise; fuzzy enhancement;
- 【文献出处】 武汉理工大学学报(信息与管理工程版) ,Journal of Wuhan University of Technology(Information & Management Engineering) , 编辑部邮箱 ,2009年02期
- 【分类号】TN912.35
- 【被引频次】5
- 【下载频次】71