DeRoom is a Machine-Learning based audio plug-in to reduce or remove reverb and room resonances in real-time. An artificial neural network has been trained on many different room scenarios in order to be able to separate direct sound from reflection components. Reverb tails and echos are automatically detected and then precisely suppressed or if desired entirely removed. DeRoom is perfectly suited to increase the quality and intelligibility of speech which has been recorded in very reverberant environments.