Non-negative Matrix Factorization (NMF) has seen growing interest in the signal processing community over the last 15 years. While initially targeted at unsupervised learning problems, the technique can also be adopted in a (semi)supervised way. In this work we aim at applying these methods for single channel speech enhancement. Here, the target speech is typically corrupted by various kinds of noise. The overall goal of the work is to improve the Signal-to-Noise ratio of different speech-noise combinations using NMF techniques. Tasks include: - literature review regarding NMF in speech signal processing - review of available test corpora (speech and noise) - review of available toolkits (Matlab) - implementation of NMF-based (semi)supervised speech enhancement algorithms (e.g. [1]) - evaluation of the implemented algorithms [1] Mohammadiha et al. (2013) Supervised and Unsupervised Speech Enhancement Using Nonnegative Matrix Factorization. IEEE Transactions on Audio, Speech, and Language Processing.