SPEECH DETECTION WITH NOISE SUPPRESSION BASED ON PRINCIPAL COMPONENTS ANALYSIS

摘要

A method for effectively suppressing background noise in a speech detection system comprises a filter bank (610) for separating source speech data into discrete frequency sub-bands to generate filtered channel energy (428), and a noise suppressor (412) for weighting the frequency sub-bands to improve the signal-to-noise ratio of the resultant noise-suppressed channel energy. The noise suppressor (412) preferably includes a subspace module (634) for using a Karhunen-Loeve transformation to create a subspace based on the background noise, a projection module (636) for generating projected channel energy by projecting the filtered channel energy (428) onto the created subspace, and a weighting module (638) for applying calculated weighting values to the projected channel energy to generate the noise-suppressed channel energy.