Discriminative Neural Clustering for Speaker Diarisation

@article{Li2021DiscriminativeNC,
  title={Discriminative Neural Clustering for Speaker Diarisation},
  author={Q. Li and Florian L. Kreyssig and Chao Zhang and P. Woodland},
  journal={2021 IEEE Spoken Language Technology Workshop (SLT)},
  year={2021},
  pages={574-581}
}
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. Com-pared 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… Expand
16 Citations
Neural Speaker Diarization with Speaker-Wise Chain Rule
  • 10
  • PDF
Optimal Mapping Loss: A Faster Loss for End-to-End Speaker Diarization
  • 3
  • PDF
Meta-Learning With Latent Space Clustering in Generative Adversarial Network for Speaker Diarization
  • 4
  • PDF
DIHARD II is Still Hard: Experimental Results and Discussions from the DKU-LENOVO Team
  • 9
  • PDF
Attention is All You Need in Speech Separation
  • 4
  • PDF
Combination of Deep Speaker Embeddings for Diarisation
  • 2
  • Highly Influenced
  • PDF
Content-Aware Speaker Embeddings for Speaker Diarisation
  • PDF
...
1
2
...

References

SHOWING 1-10 OF 44 REFERENCES
Fully Supervised Speaker Diarization
  • 100
  • PDF
Speaker diarization using deep neural network embeddings
  • 146
LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization
  • 26
  • PDF
Unsupervised Methods for Speaker Diarization: An Integrated and Iterative Approach
  • 112
  • PDF
A spectral clustering approach to speaker diarization
  • 62
  • PDF
Speaker Diarization with LSTM
  • 124
  • PDF
End-to-End Neural Speaker Diarization with Permutation-Free Objectives
  • 48
  • PDF
Deep clustering: Discriminative embeddings for segmentation and separation
  • 659
  • PDF
Speaker Diarization: A Review of Recent Research
  • 525
  • PDF
...
1
2
3
4
5
...