Informed FastICA: Semi-Blind Minimum Variance Distortionless Beamformer

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24220%2F24%3A00012440" target="_blank" >RIV/46747885:24220/24:00012440 - isvavai.cz</a>

Výsledek na webu

<a href="https://asap.ite.tul.cz/wp-content/uploads/sites/3/2024/07/IWAENC_2024___Blind_Minimum_Variance_Distortionless_Beamformer__The_Informed_FastICA_algorithm.pdf" target="_blank" >https://asap.ite.tul.cz/wp-content/uploads/sites/3/2024/07/IWAENC_2024___Blind_Minimum_Variance_Distortionless_Beamformer__The_Informed_FastICA_algorithm.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/IWAENC61483.2024.10694166" target="_blank" >10.1109/IWAENC61483.2024.10694166</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Informed FastICA: Semi-Blind Minimum Variance Distortionless Beamformer

Popis výsledku v původním jazyce

Non-Gaussianity-based Independent Vector Extraction leads to the famous one-unit FastICA/FastIVA algorithm when the likelihood function is optimized using an approximate Newton-Raphson algorithm under the orthogonality constraint. In this paper, we replace the constraint with the analytic form of the minimum variance distortionless beamformer (MVDR), by which a semi-blind variant of FastICA/FastIVA is obtained. The side information here is provided by a weighted covariance matrix replacing the noise covariance matrix, the estimation of which is a frequent goal of neural beamformers. The algorithm thus provides an intuitive connection between model-based blind extraction and learning-based extraction. The algorithm is tested in simulations and speaker ID-guided speaker extraction, showing fast convergence and promising performance.

Název v anglickém jazyce

Informed FastICA: Semi-Blind Minimum Variance Distortionless Beamformer

Popis výsledku anglicky

Non-Gaussianity-based Independent Vector Extraction leads to the famous one-unit FastICA/FastIVA algorithm when the likelihood function is optimized using an approximate Newton-Raphson algorithm under the orthogonality constraint. In this paper, we replace the constraint with the analytic form of the minimum variance distortionless beamformer (MVDR), by which a semi-blind variant of FastICA/FastIVA is obtained. The side information here is provided by a weighted covariance matrix replacing the noise covariance matrix, the estimation of which is a frequent goal of neural beamformers. The algorithm thus provides an intuitive connection between model-based blind extraction and learning-based extraction. The algorithm is tested in simulations and speaker ID-guided speaker extraction, showing fast convergence and promising performance.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10307 - Acoustics

Projekt

—

Návaznosti

V - Vyzkumna aktivita podporovana z jinych verejnych zdroju

Rok uplatnění

2024

Kód důvěrnosti údajů

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

Název statě ve sborníku

The 18th International Workshop on Acoustic Signal Enhancement (IWAENC 2024)

ISBN

979-835036185-8

ISSN

2639-4316

e-ISSN

—

Počet stran výsledku

5

Strana od-do

95-99

Název nakladatele

IEEE, Eurasip

Místo vydání

—

Místo konání akce

Aalborg, Denmark

Datum konání akce

1. 1. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001337653100020