Speaker: Alicia Lozano Díez Abstract: In this talk, we will deeply review the algorithms behind end-to-end systems for speaker diarization based on neural networks. In particular, we will describe how the encoder-decoder part of the model calculates “attractors” that capture… Read More
