Abstract
A speech processing apparatus and method may identify divisions of a signal spectrum as having a speech component or having no speech component. A comb filter is generated, based on a high-accuracy speech pitch obtained in the identified speech component divisions, for enhancing speech information in the speech component divisions. The comb filter is applied to the speech component divisions to suppress noise.
Full Text
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call