Vocal noise cancellation from respiratory sounds

摘要

Although background noise cancellation for speech or electrocardiographic recording is well established, however when the background noise contains vocal noises and the main signal is a breath sound, the problem becomes challenging as the signal and noise are no longer uncorrelated. In this paper, a technique for detecting the vocal noise within a breath sound and then removing it without hampering the main components of the breath sound is discussed. The results of this study show that the developed technique is successful and feasible in the time domain. This technique may also be used to detect and separate wheezes from breath sounds in clinical assessments.