Discriminative Training of VBx Diarization

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F24%3APU152297" target="_blank" >RIV/00216305:26230/24:PU152297 - isvavai.cz</a>

Result on the web

<a href="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446119" target="_blank" >https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10446119</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICASSP48485.2024.10446119" target="_blank" >10.1109/ICASSP48485.2024.10446119</a>

Result language

angličtina

Original language name

Discriminative Training of VBx Diarization

Original language description

Bayesian HMM clustering of x-vector sequences (VBx) has be- come a widely adopted diarization baseline model in publications and challenges. It uses an HMM to model speaker turns, a gen- eratively trained probabilistic linear discriminant analysis (PLDA) for speaker distribution modeling, and Bayesian inference to esti- mate the assignment of x-vectors to speakers. This paper presents a new framework for updating the VBx parameters using discrim- inative training, which directly optimizes a predefined loss. We also propose a new loss that better correlates with the diarization error rate compared to binary cross-entropy - the default choice for diarization end-to-end systems. Proof-of-concept results across three datasets (AMI, CALLHOME, and DIHARD II) demonstrate the method's capability of automatically finding hyperparameters, achieving comparable performance to those found by extensive grid search, which typically requires additional hyperparameter behavior knowledge. Moreover, we show that discriminative fine-tuning of PLDA can further improve the model's performance. We release the source code with this publication.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

Result was created during the realization of more than one project. More information in the Projects tab.

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2024

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ISBN

979-8-3503-4485-1

ISSN

—

e-ISSN

—

Number of pages

5

Pages from-to

11871-11875

Publisher name

IEEE Signal Processing Society

Place of publication

Seoul

Event location

Seoul

Event date

Apr 14, 2024

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—