Voice Signal Analysis and Processing Course Design with MATLAB Implementation

In this course design project, we will perform voice signal analysis and processing using MATLAB. The implementation begins with recording your personal voice signal and applying sampling to the recorded audio data. The sampled signal's time-domain waveform and spectrogram will be plotted using MATLAB's plotting functions like plot() for time-domain visualization and spectrogram() for frequency-domain analysis. To meet specified filter performance indicators, you will design digital filters employing either the window function method (using functions like fir1() with various windows) or bilinear transformation method (utilizing butter() or cheby1() for IIR filter design). The filter's frequency response will be visualized through freqz() function plots. The designed filter will then be applied to the acquired voice signal using MATLAB's filter() function for convolution-based filtering. Post-filtering, we will generate the filtered signal's time-domain waveform and spectrogram for comparative analysis. A detailed comparison between pre-filtering and post-filtering signals will be conducted to analyze signal changes in both time and frequency domains, examining parameters like amplitude variations, frequency component modifications, and signal-to-noise ratio improvements. Finally, the processed audio signal will be played back using MATLAB's sound() or audiowrite() functions to audibly demonstrate the filtering effects.