With the increasing use of hands-free telephony, especially within cars, it is often the case that speech signals are contaminated by the addition of unwanted background acoustic noise. The goal of a speech enhancer is to reduce or eliminate this background noise without distorting the speech signal. Although current speech enhancers are able to reduce the perceived noise it has been found that they invariably also make the speech less intelligible. Recent research has indicated that intelligibility depends on the faithful reproduction of the modulation of the spectral amplitudes and this is not preserved by current speech enhancers. This project will develop statistical models of the modulation-domain characteristics of speech and noise signals. It will then use these to define an optimized speech enhancement algorithm that does not degrade intelligibility.
Researchers: Yu Wang and Mike Brookes
- Speech enhancement using a robust Kalman filter post-processor in the modulation domain. In: Proc. IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP), Vancouver, Canada, 2013.(2013)
- A Subpace Method for Speech Enhancement in the Modulation Domain. In: Proc. European Signal Processing Conference (EUSIPCO), Marrakech, Morocco, 2013.(2013)