Target Speech Extraction: Independent Vector Extraction Guided by Supervised Speaker Identification

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24220%2F22%3A00009853" target="_blank" >RIV/46747885:24220/22:00009853 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/document/9829828" target="_blank" >https://ieeexplore.ieee.org/document/9829828</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/TASLP.2022.3190739" target="_blank" >10.1109/TASLP.2022.3190739</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Target Speech Extraction: Independent Vector Extraction Guided by Supervised Speaker Identification

Popis výsledku v původním jazyce

This manuscript proposes a novel robust procedure for the extraction of a speaker of interest (SOI) from a mixture of audio sources. The estimation of the SOI is performed via independent vector extraction (IVE). Since the blind IVE cannot distinguish the target source by itself, it is guided towards the SOI via frame-wise speaker identification based on deep learning. Still, an incorrect speaker can be extracted due to guidance failings, especially when processing challenging data. To identify such cases, we propose a criterion for non-intrusively assessing the estimated speaker. It utilizes the same model as the speaker identification, so no additional training is required. When incorrect extraction is detected, we propose a ``deflation‘‘ step in which the incorrect source is subtracted from the mixture and, subsequently, another attempt to extract the SOI is performed. The process is repeated until successful extraction is achieved. The proposed procedure is experimentally tested on artificial and real-world datasets containing challenging phenomena: source movements, reverberation, transient noise, or microphone failures. The method is compared with state-of-the-art blind algorithms as well as with current fully supervised deep learning-based methods.

Název v anglickém jazyce

Target Speech Extraction: Independent Vector Extraction Guided by Supervised Speaker Identification

Popis výsledku anglicky

This manuscript proposes a novel robust procedure for the extraction of a speaker of interest (SOI) from a mixture of audio sources. The estimation of the SOI is performed via independent vector extraction (IVE). Since the blind IVE cannot distinguish the target source by itself, it is guided towards the SOI via frame-wise speaker identification based on deep learning. Still, an incorrect speaker can be extracted due to guidance failings, especially when processing challenging data. To identify such cases, we propose a criterion for non-intrusively assessing the estimated speaker. It utilizes the same model as the speaker identification, so no additional training is required. When incorrect extraction is detected, we propose a ``deflation‘‘ step in which the incorrect source is subtracted from the mixture and, subsequently, another attempt to extract the SOI is performed. The process is repeated until successful extraction is achieved. The proposed procedure is experimentally tested on artificial and real-world datasets containing challenging phenomena: source movements, reverberation, transient noise, or microphone failures. The method is compared with state-of-the-art blind algorithms as well as with current fully supervised deep learning-based methods.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

20201 - Electrical and electronic engineering

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název periodika

IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING

ISSN

2329-9290

e-ISSN

—

Svazek periodika

30

Číslo periodika v rámci svazku

30

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

15

Strana od-do

2295-2309

Kód UT WoS článku

000831126700006

EID výsledku v databázi Scopus

2-s2.0-85135228612