针对语音通话质量的提升问题,提出一种基于快速傅里叶变换(Fast Fourier Transform,FFT)和Masking技术的实时语音通话降噪算法.首先,提出一个实时语音通话降噪的基本框架,并研究了帧分割、窗函数处理及FFT的数学原理.其次,阐述了基于人耳听觉特性的Masking方法及其在频域中的应用.最后,通过逆快速傅里叶变换(Inverse Fast Fourier Transform,IFFT)将信号转换回时域,并进行实验分析.实验结果表明,该降噪算法可以有效改善语音的清晰度和整体感知质量.
Real-Time Voice Call Denoising Algorithm Based on FFT and Masking
A real-time voice call denoising algorithm based on Fast Fourier Transform (FFT) and Masking technology is proposed to improve the quality of voice calls. Firstly, a basic framework for real-time voice call denoising is proposed, and the mathematical principles of frame segmentation, window function processing, and FFT are studied. Secondly, the Masking method based on human auditory characteristics and its application in the frequency domain were elaborated. Finally, the signal is converted back to the time domain through Inverse Fast Fourier Transform (IFFT) and subjected to experimental analysis. The experimental results show that the denoising algorithm can effectively improve the clarity and overall perceptual quality of speech.
NETL
NSTL
万方数据
