Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx) for Combined End-to-End and Vector Clustering-based Diarization

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F23%3APU149716" target="_blank" >RIV/00216305:26230/23:PU149716 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.isca-speech.org/archive/pdfs/interspeech_2023/delcroix23_interspeech.pdf" target="_blank" >https://www.isca-speech.org/archive/pdfs/interspeech_2023/delcroix23_interspeech.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.21437/Interspeech.2023-628" target="_blank" >10.21437/Interspeech.2023-628</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx) for Combined End-to-End and Vector Clustering-based Diarization

Popis výsledku v původním jazyce

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.

Název v anglickém jazyce

Multi-Stream Extension of Variational Bayesian HMM Clustering (MS-VBx) for Combined End-to-End and Vector Clustering-based Diarization

Popis výsledku anglicky

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.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/GX19-26934X" target="_blank" >GX19-26934X: Neuronové reprezentace v multimodálním a mnohojazyčném modelování</a><br>

Návaznosti

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

Rok uplatnění

2023

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

ISBN

—

ISSN

1990-9772

e-ISSN

—

Počet stran výsledku

5

Strana od-do

3477-3481

Název nakladatele

International Speech Communication Association

Místo vydání

Dublin

Místo konání akce

Dublin

Datum konání akce

20. 8. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—