Speech

Echo Cancellation
Acoustic Echo Cancellation deals with the question how to eliminate an occurring echo in conferencing or hands-free communication situations. Key element is the estimation of the acoustical path, over which the incoming signal is transmitted and afterwards received at the microphone as a disturbing echo signal. Therefore adaptive algorithms are used and combined with a postfilter for residual echo supression.
Multi-Microphone Systems
Reverberant signals cause problems. Dereverberation algorithms suppress reverberation. Typical algorithms and their applications are introduced.
Bandwidth Extension
"Artificial Bandwidth Extension" adds synthetic frequency components to narrowband speech. Thus a more natural sounding speech reproduction and an increased intelligibility is achieved.
Intelligibility Enhancement
Algorithms for near-end listening enhancement (NELE) improve the intelligibility of speech from the far-end, played back in a noisy near-end environment, by adaptively filtering the speech signal and taking into account the near-end background noise characteristics.
Dereverberation
Dereverberation algorithms remove unwanted reverberation from signals recorded in reverberant rooms. Typical applications of dereverberation are the improvement of telephone or hearing aid speech quality or automatic speech recognition.
Noise Reduction
Given a noisy input signal, noise reduction aims at attenuates the noise signal component as much as possible while keeping the distortion of the underlying speech signal as low as possible.