基于时频分析方法LPFT的语音处理GUI系统

由于语音的非平稳性及时变性,时频分析方法是处理语音信号的重要工具.然而线性短时傅里叶变换的时频聚集性较差,而双线性维格纳变换在处理多分量信号时会受到交叉项的干扰.为了克服以上两种时频分析方法的缺点,利用短时傅里叶变换的扩展形式即局部多项式傅里叶变换LPFT来处理语音信号,建立了基于LPFT的语音处理GUI系统,实现了在时域、频域和时频域对语音的分析和对比.并给出语音处理的例子,验证了LPFT方法与其它方法相比所具有的优势.该系统简明直观,是语音处理的较好的平台.

Speech Analysis GUI System Based on the Local Polynomial Fourier Transform

Since the speech is nonstationary and time-varying, the time-frequency analysis is a useful tool to process the speech. However, the linear short-time Fourier transform (STFT) suffers from low resolution in the time-frequency domain, and the bilinear Wigner-Ville distribution (WVD) suffers from the cross terms when processing multi-component signals. Therefore, in this paper, the generalization of the STFT, that is, the local polynomial Fourier transform (LPFT) is employed to process the speech, and a speech analysis GUI system based on the LPFT is proposed. In the proposed GUI system, the analysis and comparisons for speech are realized in the time domain, the frequency domain, and the time-frequency domain. It is shown by the example that the LPFT is more suitable to process the speech. The proposed system is user-friendly, and is a good platform to process the speech.