MATLAB中实现的两阶段法语音去混响算法。

简介：
The directory houses MATLAB codes that realize a reverberant speech enhancement algorithm, as presented in Mingyang Wu and DeLiang Wang’s work, “Two-stage algorithm for enhancement of reverberant speech,” published at the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) in 2005, and also appearing in “Two-stage Algorithm for One-microphone Reverberant Speech Enhancement” within the July 2006 issue of IEEE Transactions on Speech and Audio Processing. Furthermore, it includes a sample program named sample.m designed to load and process the data, subsequently generating and displaying the outcomes. The directory also contains a raw data file labeled data, alongside a data file named sample.mat which holds the original speech data, as well as a WAV file designated wav representing the processed results. Additionally, there are separate WAV files: org.wav containing clean speech, rev.wav representing reverberant speech, inv.wav illustrating inverse filtered speech, and “derev.wav” showcasing the ultimately processed audio signal. Finally, a readme text file is present for documentation purposes.