Utterance Clustering Using Stereo Audio Channels

Utterance clustering is one of the actively researched topics in audio signal processing and machine learning. This study aims to improve performance utterance by multichannel (stereo) signals. Processed signals were generated combining left- right-channel a few different ways then extracting embedded features (also called d-vectors) from those processed applied Gaussian mixture model for supervised clustering. In training phase, parameter-sharing was obtained train each speaker. testing speaker with maximum likelihood selected as detected Results experiments real recordings multiperson discussion sessions showed that proposed method used achieved significantly better than conventional mono-audio more complicated conditions.