Combining the glottal mixture model (GLOMM) with UBM for speaker recognition


We present an iterative algorithm to extract the voice source waveform from recordings of speech for speaker identification. The method detects glottal closings, then constructs a speaker-dependent library of glottal pulse waveforms by clustering data windows centered on the linear prediction error time-series at the glottal closures. With the voice source… (More)
DOI: 10.1109/EUSIPCO.2016.7760630