Explore MATLAB source code curated for "语音信号" with clean implementations, documentation, and examples.
MATLAB implementation for automatic speech onset and offset detection utilizing short-time energy and zero-crossing rate features of speech signals, with effective performance and practical voice processing applications.
A short-term energy analysis program for speech signals enabling voiced/unvoiced segmentation with frame-based processing and energy thresholding algorithms
Implementing frame segmentation on sampled speech signals and storing the framed signals as a matrix for further analysis
A comprehensive MATLAB-based filter analysis of speech signals, demonstrating practical implementation techniques including spectral analysis, time-domain processing, and frequency component extraction using built-in filter design functions - highly beneficial for learning digital signal processing applications.
Improved signal matching pursuit sparse decomposition algorithm based on Gabor time-frequency atoms, demonstrating superior speech signal reconstruction performance with optimized feature extraction and computational efficiency.
This project involves recording a personal speech signal, performing sampling operations, and visualizing both time-domain waveforms and spectrograms. With specified filter performance metrics, we design filters using the window function method and bilinear transformation, then plot their frequency responses. The recorded signal is filtered using the custom-designed filter, and the filtered signal's time-domain waveform and spectrum are visualized. Finally, a comparative analysis between pre-filtered and post-filtered signals is conducted to examine signal changes.
MATLAB implementation of Kalman filtering for enhancing speech signals corrupted by white Gaussian noise, featuring effective noise reduction and signal restoration capabilities with satisfactory performance results.
Blind Source Separation (BSS) is a technique for recovering independent source signals from mixed observations without prior knowledge of the mixing process or source characteristics. This method proves particularly effective for extracting individual speech signals from multi-microphone recordings. This article provides detailed explanations of LMS and RLS algorithms with code implementation insights, conducts comparative performance analysis through simulations, and demonstrates practical speech signal separation using these approaches.
Spectral Analysis of Speech Signals, including FFT transformation of sampled speech and cepstrum analysis with code implementation insights.
MATLAB-based linear predictive coding (LPC) analysis for speech signals, featuring a complete GUI interface implementation and including test datasets for validation.