Spherediar: An Effective Speaker Diarization System for Meeting Data

Kaseva, Tuomas; Rouhe, Aku; Kurimo, Mikko

Aaltodoc
→
6b Artikkelit / Articles from Research information portal
→
[article-cris] Sähkötekniikan korkeakoulu / ELEC
→
View Item

Spherediar: An Effective Speaker Diarization System for Meeting Data

Altmetrics



Title:	Spherediar: An Effective Speaker Diarization System for Meeting Data
Author(s):	Kaseva, Tuomas ; Rouhe, Aku ; Kurimo, Mikko
Date:	2019-12-01
Language:	en
Pages:	8 373-380
Department:	Dept Signal Process and Acoust
ISBN:	9781728103068
Series:	2019 IEEE Automatic Speech Recognition and Understanding Workshop, ASRU 2019 - Proceedings
DOI-number:	10.1109/ASRU46091.2019.9003967
Keywords:	segmentation, silhouette coefficients, speaker diarization, speaker embeddings, spherical K-means
» Show full item record
This publication is imported from Aalto University Research information portal: https://research.aalto.fi > View this publication in Research information portal > Other link related to publication (Research information portal) > View/Open full text file from Research information portal
Citation: Kaseva , T , Rouhe , A & Kurimo , M 2019 , Spherediar: An Effective Speaker Diarization System for Meeting Data . in 2019 IEEE Automatic Speech Recognition and Understanding Workshop, ASRU 2019 - Proceedings . , 9003967 , IEEE , pp. 373-380 , IEEE Automatic Speech Recognition and Understanding Workshop , Singapore , Singapore , 15/12/2019 . https://doi.org/10.1109/ASRU46091.2019.9003967
Abstract: In this paper, we present SphereDiar, a speaker diarization system composed of three novel subsystems: The Sphere-Speaker (SS) neural network, designed for speaker embedding extraction, a segmentation method called Homogeneity Based Segmentation (HBS) and a clustering algorithm called Top Two Silhouettes (Top2S). The system is evaluated on a set of over 200 manually transcribed multiparty meetings. The evaluation reveals that the system can be further simplified by omitting the use of HBS. Furthermore, we illustrate that SphereDiar achieves state-of-The-Art results with two different meeting data sets.

Description: \| openaire: EC/H2020/780069/EU//MeMAD
Permanent link to this item: http://urn.fi/URN:NBN:fi:aalto-202004092797 Email this Export to RefWorks QR Code Print BibTex Tweet