Publication Details
Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx) for Combined End-to-End and Vector Clustering-based Diarization
TAWARA, N.
DIEZ SÁNCHEZ, M.
Landini Federico Nicolás, Ph.D. (RG SPEECH)
Silnova Anna, M.Sc., Ph.D. (DCGM)
Ogawa Atsunori
Nakatani Tomohiro
Burget Lukáš, doc. Ing., Ph.D. (DCGM)
ARAKI, S.
speaker diarization, end-to-end, VBx, clustering
Combining end-to-end neural speaker diarization (EEND) with vector clustering
(VC), known as EEND-VC, has gained interest for leveraging the strengths of both
methods. EEND-VC estimates activities and speaker embeddings for all speakers
within an audio chunk and uses VC to associate these activities with speaker
identities across different chunks. EEND-VC generates thus multiple streams of
embeddings, one for each speaker in a chunk. We can cluster these embeddings
using constrained agglomerative hierarchical clustering (cAHC), ensuring
embeddings from the same chunk belong to different clusters. This paper
introduces an alternative clustering approach, a multi-stream extension of the
successful Bayesian HMM clustering of x-vectors (VBx), called MS-VBx. Experiments
on three datasets demonstrate that MS-VBx outperforms cAHC in diarization and
speaker counting performance.
@inproceedings{BUT185573,
author="DELCROIX, M. and TAWARA, N. and DIEZ SÁNCHEZ, M. and LANDINI, F. and SILNOVA, A. and OGAWA, A. and NAKATANI, T. and BURGET, L. and ARAKI, S.",
title="Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx) for Combined End-to-End and Vector Clustering-based Diarization",
booktitle="Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",
year="2023",
journal="Proceedings of Interspeech",
volume="2023",
number="08",
pages="3477--3481",
publisher="International Speech Communication Association",
address="Dublin",
doi="10.21437/Interspeech.2023-628",
issn="1990-9772",
url="https://www.isca-speech.org/archive/pdfs/interspeech_2023/delcroix23_interspeech.pdf"
}