BUT System Description for DIHARD Speech Diarization Challenge 2019

TL;DR

This paper details the BUT team's speech diarization systems for the DIHARD 2019 challenge, utilizing clustering and HMM techniques across multiple tracks to improve speaker segmentation accuracy.

Contribution

It introduces a multi-track diarization approach combining AHC, x-vectors, and Bayesian HMMs, tailored for different challenge tracks.

Findings

Effective clustering and HMM integration improved diarization performance.

Systems achieved competitive results in DIHARD 2019 challenge.

Multi-channel x-vector extraction enhanced speaker segmentation.

Abstract

This paper describes the systems developed by the BUT team for the four tracks of the second DIHARD speech diarization challenge. For tracks 1 and 2 the systems were based on performing agglomerative hierarchical clustering (AHC) over x-vectors, followed by the Bayesian Hidden Markov Model (HMM) with eigenvoice priors applied at x-vector level followed by the same approach applied at frame level. For tracks 3 and 4, the systems were based on performing AHC using x-vectors extracted on all channels.