Multi-Speaker and Wide-Band Simulated Conversations as Training Data for End-to-End Neural Diarization

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F23%3APU149424" target="_blank" >RIV/00216305:26230/23:PU149424 - isvavai.cz</a>

Result on the web

<a href="https://ieeexplore.ieee.org/document/10097049" target="_blank" >https://ieeexplore.ieee.org/document/10097049</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICASSP49357.2023.10097049" target="_blank" >10.1109/ICASSP49357.2023.10097049</a>

Result language

angličtina

Original language name

Multi-Speaker and Wide-Band Simulated Conversations as Training Data for End-to-End Neural Diarization

Original language description

End-to-end diarization presents an attractive alternative to standard cascaded diarization systems because a single system can handle all aspects of the task at once. Many flavors of end-to-end models have been proposed but all of them require (so far non-existing) large amounts of annotated data for training. The compromise solution consists in generating synthetic data and the recently proposed simulated conversations (SC) have shown remarkable improvements over the original simulated mixtures (SM). In this work, we create SC with multiple speakers per conversation and show that they allow for substantially better performance than SM, also reducing the dependence on a fine-tuning stage. We also create SC with wide-band public audio sources and present an analysis on several evaluation sets. Together with this publication, we release the recipes for generating such data and models trained on public sets as well as the implementation to efficiently handle multiple speakers per conversation and an auxiliary voice activity detection loss.

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

2023

Confidentiality

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

Article name in the collection

Proceedings of ICASSP 2023

ISBN

978-1-7281-6327-7

ISSN

—

e-ISSN

—

Number of pages

5

Pages from-to

1-5

Publisher name

IEEE Signal Processing Society

Place of publication

Rhodes Island

Event location

Rhodes Island, Greece

Event date

Jun 4, 2023

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—