LGCLSDASSep 22, 2021

Diarisation using location tracking with agglomerative clustering

arXiv:2109.10598v2
Originality Incremental advance
AI Analysis

This addresses speaker diarisation for meeting transcription by relaxing stationarity assumptions, but it is incremental as it builds on existing clustering methods with location information.

The paper tackled the problem of speaker diarisation in meetings where speakers move, by modeling speaker movements with Kalman filters within an agglomerative clustering framework, resulting in improvements on a Microsoft transcription task compared to methods without location tracking or stationarity assumptions.

Previous works have shown that spatial location information can be complementary to speaker embeddings for a speaker diarisation task. However, the models used often assume that speakers are fairly stationary throughout a meeting. This paper proposes to relax this assumption, by explicitly modelling the movements of speakers within an Agglomerative Hierarchical Clustering (AHC) diarisation framework. Kalman filters, which track the locations of speakers, are used to compute log-likelihood ratios that contribute to the cluster affinity computations for the AHC merging and stopping decisions. Experiments show that the proposed approach is able to yield improvements on a Microsoft rich meeting transcription task, compared to methods that do not use location information or that make stationarity assumptions.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes