Self-Learning Speaker Identification door Tobias Herbig & Franz Gerl