LPC POLES TRACKER FOR MUSIC/SPEECH/NOISE SEGMENTATION AND MUSIC CANCELLATION

Abstract

In automatic speech recognition (ASR) of broadcast news shows the input utterances are often corrupted by background music and noise. This paper proposes a new method of automatic segmentation a speech signals according to the background: music, clean or noisy. LPC analysis is used to extract the poles of the associated transfer function. Based on the time evolution of the poles it is possible to discriminate the contributions of music, speech and noise: music poles are stabler longer than speech poles while noise poles have a more unstable behavior than speech poles. Once the background of a signal is identified, poles tagged as non-speech can be separated from speech poles. Using only the speech poles along with the LPC residuals, it is possible to reconstruct a new signal freed of music and noise contributions.