Discriminative Neural Clustering For Speaker Diarisation

Abstract

In this paper, we propose Discriminative Neural Clustering (DNC) that formulates data clustering with a maximum number of clusters as a supervised sequence-to-sequence learning problem. Compared to traditional unsupervised clustering algorithms, DNC learns clustering patterns from training data without requiring an explicit definition of a similarity measure. An implementation of DNC based on the Transformer architecture is shown to be effective on a speaker diarisation task using the challenging AMI dataset. Since AMI contains only 147 complete meetings as individual input sequences, data scarcity is a significant issue for training a Transformer model for DNC. Accordingly, this paper proposes three data augmentation schemes: sub-sequence randomisation, input vector randomisation, and Diaconis augmentation, which generates new data samples by rotating the entire input sequence of L2-normalised speaker embeddings. Experimental results on AMI show that DNC achieves a reduction in speake

Stats

• citations21
• S2 citations—
• github stars0
• HF likes0
• heat score10.07
• arxiv keyli2019discriminative

Related papers

• Deep Self-supervised Hierarchical Clustering For Speaker Diarization (2020)5.24
• Meta-learning With Latent Space Clustering In Generative Adversarial Network For Speaker Diarization (2020)9.03
• Enhancements For Audio-only Diarization Systems (2019)0.00
• Advances In Integration Of End-to-end Neural And Clustering-based Diarization For Real Conversational Speech (2021)16.48
• Spectral Clustering-aware Learning Of Embeddings For Speaker Diarisation (2022)2.26
• Turn-to-diarize: Online Speaker Diarization Constrained By Transformer Transducer Speaker Turn Detection (2021)12.40
• Learning Deep Representations By Multilayer Bootstrap Networks For Speaker Diarization (2019)0.00
• Interrelate Training And Searching: A Unified Online Clustering Framework For Speaker Diarization (2022)6.77