Abstract: Presents an approach to modeling speech spectra and pitch for text-independent speaker identification using Gaussian mixture models based on multi-space probability distribution (MSD-GMM).
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback