Search - compare d

Search - compare d - List

[Speech/Voice recognition/combine] 19_M-W-K-D-D_SPA2010_final DL : 0: This paper presents results of speaker recognition experiments using short Polish sentences. We developed and analyzed various vector quantization representations in order to first maximize identification effectiveness and second to compare VQ (vector quantization) and GMM (Gaussian mixture model) approaches. For the research and experiments we created and exploited database, containing specially prepared short speech sequences.
Date : 2025-12-26 Size : 252kb User : Tomasz
[Speech/Voice recognition/combine] chenxu DL : 0: （1）录制一段语音信号，完成对信号的采样，画出信号的时域波形和频谱图，确定信号的频谱范围；（2）给信号叠加噪声（噪声类型分为如下几种：a白噪声；b单频噪色（正弦干扰）；c多频噪声（多正弦干扰）；d其它干扰。），画出受噪声干扰的信号时域波形和频谱图；（3）采用窗函数法设计FIR低通滤波器，画出滤波器的频响特性图；（4）用所设计的滤波器对受噪声影响的信号进行滤波，画出滤波后语音信号的时域波形图和频谱图；（5）对滤波前后的信号进行对比，分析信号的变化；回放语音信号，并与原始语音信号对比。((1) record a speech signal, complete the sampling of the signal, draw the time domain waveform and frequency spectrum of the signal, and determine the spectrum range of the signal; (2) signal superposition noise (noise type is divided into the following: a white noise, B single frequency noise (sinusoidal interference); C multifrequency noise (multi sinusoidal interference); d other interference) Noise signal, time domain waveform and spectrogram are drawn; (3) the FIR low-pass filter is designed by using the window function method, and the frequency response characteristic of the filter is drawn; (4) the designed filter is used to filter the signal affected by the noise, and the time domain waveform and spectrogram of the filtered speech signal are drawn; (5) compare the signals before and after filtering, analyze the change of the signal, replay the speech signal, and compare with the original speech signal)
Date : 2025-12-26 Size : 13kb User : 华