Block-online Multi-channel Speech Enhancement Using Deep Neural Network-supported Relative Transfer Function Estimates

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24220%2F20%3A00006552" target="_blank" >RIV/46747885:24220/20:00006552 - isvavai.cz</a>

Result on the web

<a href="https://digital-library.theiet.org/content/journals/10.1049/iet-spr.2019.0304" target="_blank" >https://digital-library.theiet.org/content/journals/10.1049/iet-spr.2019.0304</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1049/iet-spr.2019.0304" target="_blank" >10.1049/iet-spr.2019.0304</a>

Result language

angličtina

Original language name

Block-online Multi-channel Speech Enhancement Using Deep Neural Network-supported Relative Transfer Function Estimates

Original language description

This work addresses the problem of block-online processing for multi-channel speech enhancement. Such processing is vital in scenarios with moving speakers and/or when very short utterances are processed, e.g., in voice assistant scenarios. We consider several variants of a system that performs beamforming supported by DNN-based voice activity detection (VAD) followed by post-filtering. The speaker is targeted through estimating relative transfer functions between microphones. Each block of the input signals is processed independently in order to make the method applicable in highly dynamic environments. Owing to the short length of the processed block, the statistics required by the beamformer are estimated less precisely. The influence of this inaccuracy is studied and compared to the processing regime when recordings are treated as one block (batch processing). The experimental evaluation of the proposed method is performed on large datasets of CHiME-4 and on another dataset featuring moving target speaker. The experiments are evaluated in terms of objective and perceptual criteria (such as signal-to-interference ratio (SIR) or perceptual evaluation of speech quality (PESQ), respectively). Moreover, word error rate (WER) achieved by a baseline automatic speech recognition system is evaluated, for which the enhancement method serves as a front-end solution. The results indicate that the proposed method is robust with respect to short length of the processed block. Significant improvements in terms of the criteria and WER are observed even for the block length of 250 ms.

Czech name

—

Czech description

—

Type

J_imp - Article in a specialist periodical, which is included in the Web of Science database

CEP classification

—

OECD FORD branch

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

Project

<a href="/en/project/GA17-00902S" target="_blank" >GA17-00902S: Advanded Joint Blind Source Separation Methods</a><br>

Continuities

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

Publication year

2020

Confidentiality

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

Name of the periodical

IET Signal Processing

ISSN

1751-9675

e-ISSN

—

Volume of the periodical

14

Issue of the periodical within the volume

3

Country of publishing house

GB - UNITED KINGDOM

Number of pages

10

Pages from-to

124-133

UT code for WoS article

000530448800002

EID of the result in the Scopus database

2-s2.0-85084193404