The multichannel audio source separation evaluation framework is designed to facilitate the development and evaluation of audio source separation algorithms. The framework generates the mixture(s), provides the mixture(s) to the separation algorithm(s), and evaluates the outputs of the separation algorithm(s). The framework can also calculate and evaluate the ideal masks for the purposes of comparison.

Sources may have any number of channels; the framework evaluates each channel. The use of iosr.bss.mixture objects facilitate the evaluation of spatialised mixtures (e.g. binaural).


Manager: Christopher Hummersone
Developer: Joachim Fainberg