Bayesian HMM clustering of x-vector sequences (VBx) in speaker diarization: theory, implementation and analysis on standard tasks

TL;DR

This paper introduces a Bayesian HMM clustering method for speaker diarization using x-vectors, demonstrating its superior performance on standard datasets and providing new derivations, implementation details, and evaluation protocols.

Contribution

It presents the first derivation and efficient implementation of the VBx model, along with state-of-the-art diarization results and a new evaluation protocol for AMI data.

Findings

VBx outperforms other diarization methods on CALLHOME, AMI, and DIHARDII datasets.

The paper provides the derivation and update formulas for the VBx model.

New standardized evaluation protocol for AMI dataset proposed.

Abstract

The recently proposed VBx diarization method uses a Bayesian hidden Markov model to find speaker clusters in a sequence of x-vectors. In this work we perform an extensive comparison of performance of the VBx diarization with other approaches in the literature and we show that VBx achieves superior performance on three of the most popular datasets for evaluating diarization: CALLHOME, AMI and DIHARDII datasets. Further, we present for the first time the derivation and update formulae for the VBx model, focusing on the efficiency and simplicity of this model as compared to the previous and more complex BHMM model working on frame-by-frame standard Cepstral features. Together with this publication, we release the recipe for training the x-vector extractors used in our experiments on both wide and narrowband data, and the VBx recipes that attain state-of-the-art performance on all three datasets. Besides, we point out the lack of a standardized evaluation protocol for AMI dataset and we propose a new protocol for both Beamformed and Mix-Headset audios based on the official AMI partitions and transcriptions.