A computationally efficient and distributed speaker diarization framework for networked IoT-style audio devices using Federated Learning.