Low-latency online speaker diarization with graph-based label generation

Y Zhang, Q Lin, W Wang, L Yang, X Wang… - arXiv preprint arXiv …, 2021 - arxiv.org
… introduces an online speaker diarization sys… online fashion by introducing a label matching
algorithm. This algorithm solves the inconsistency between output labels and hidden labels

Self-supervised metric learning with graph clustering for speaker diarization

P Singh, S Ganapathy - 2021 IEEE Automatic Speech …, 2021 - ieeexplore.ieee.org
… novel algorithm for speaker diarization using metric learning for graph based clustering. The
… The initialization is a critical step for self-supervised training to generate reliable labels. We …

Overlap-aware End-to-End Supervised Hierarchical Graph Clustering for Speaker Diarization

P Singh, S Ganapathy - arXiv preprint arXiv:2401.12850, 2024 - arxiv.org
labeled data. In this work, we describe one of the first efforts in exploring hierarchical graph
clustering frameworks for speaker diarization… first step is the creation of a graph based on the …

[PDF][PDF] Graph Clustering Approaches for Speaker Diarization of Conversational Speech

P Singh - 2023 - leap.ee.iisc.ac.in
… clustering (PIC), a graph-based clustering algorithm. The PIC is … field of speaker diarization
has focused on generating robust … This approach allows to predict speaker labels in an online

Turn-to-diarize: Online speaker diarization constrained by transformer transducer speaker turn detection

W Xia, H Lu, Q Wang, A Tripathi… - ICASSP 2022-2022 …, 2022 - ieeexplore.ieee.org
… In this paper, we present a novel speaker diarization system … the speaker turns, represent
each speaker turn by a speaker … annotations of time-stamped speaker labels for training, our …

A Review of Common Online Speaker Diarization Methods

R Aperdannier, S Schacht, A Piazza - arXiv preprint arXiv:2406.14464, 2024 - arxiv.org
… This type of speaker diarization is known as online speaker diarization. … Li, “Low-Latency
Online Speaker Diarization with Graph-Based Label Generation,” Jun. 2022, arXiv:2111.13803 […

Online neural diarization of unlimited numbers of speakers using global and local attractors

S Horiguchi, S Watanabe, P García… - … on Audio, Speech …, 2022 - ieeexplore.ieee.org
diarization by formulating it as a multilabel classification … perform speaker diarization of an
unseen number of speakers … into short blocks after generating frame-wise embeddings from …

End-to-end Online Speaker Diarization with Target Speaker Tracking

W Wang, M Li - arXiv preprint arXiv:2310.08696, 2023 - arxiv.org
… deviation over the feature dimension, thus generating a vector of dimension 2C for each
frame… Since we already know the order of the speakers in the ground truth label Y ∈ {0, 1}T ×N , …

[PDF][PDF] Online Speaker Diarization with Core Samples Selection.

Y Yue, J Du, MK He, YT Yeung, R Wang - INTERSPEECH, 2022 - isca-archive.org
… Finally, we solve the label ambiguity problem by a global … graph-based reclustering
process [23] was also designed to improve the performance of chkpt-AHC online diarization

An Online Diarization Approach for Streaming Applications Based on Tree-Clustering and Bayesian Resegmentation

JM Martín-Doñas, H Arzelus, A Álvarez… - … Conference on Text …, 2023 - Springer
… This paper describes our proposed system for online speaker diarization suitable for … A
low-latency graph-based label generation is described in [12]. This approach modifies an …