End-to-End Diarization utilizing Attractor Deep Clustering
Abstract
Speaker diarization remains challenging due to the need for structured speaker representations, efficient modeling, and robustness to varying conditions. We propose a performant, compact diarization framework that integrates conformer decoders, transformer-updated attractors, and a deep clustering style angle loss. Our approach refines speaker representations with an enhanced conformer structure, incorporating cross-attention to attractors and an additional convolution module. To enforce structured embeddings, we extend deep clustering by constructing label-attractor vectors, aligning their directional structure with audio embeddings. We also impose orthogonality constraints on active attractors for better speaker separation while suppressing non-active attractors to prevent false activations. Finally, a permutation invariant training binary cross-entropy loss refines speaker detection. Experiments show that our method achieves low diarization error while maintaining parameter count.
Cite
@article{arxiv.2506.11090,
title = {End-to-End Diarization utilizing Attractor Deep Clustering},
author = {David Palzer and Matthew Maciejewski and Eric Fosler-Lussier},
journal= {arXiv preprint arXiv:2506.11090},
year = {2025}
}
Comments
To appear at INTERSPEECH 2025