Single-channel noise reduction schemes are designed to enhance noisy speech signals by making use of a single microphone only. This type of noise reduction therefore has to rely entirely on temporal, spectral and statistical differences between the speech signal and the background noise, and cannot exploit spatial differences as is commonly done in multi-microphone approaches. Single-channel noise reduction schemes are known to be most effective against diffuse noise sources. Many implementations have been derived during the past years. Some of them are commonly used nowadays in hearing aid and cochlear implant devices. In this project we look at a number of more advanced, recently developed variants.