摘要 |
<p>A method for implementing a noise suppressor (412) in a speech recognition system comprises a filter bank (610) for separating source speech data into discrete frequency sub-bands to generate filtered channel energy, 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 supppressor (412) preferably includes a noise calculator (634) for calculating background noise values, a speech energy calculator (636) for calculating speech energy values for each channel of the filter bank (610), and a weighting module (638) for applying calculated weighting values to the projected channel energy to generate the noise-suppressed channel energy.</p> |