Neural Directional Filtering: Far-Field Directivity Control With a Small Microphone Array

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wechsler, Julian, Chetupalli, Srikanth Raj, Halimeh, Mhd Modar, Thiergart, Oliver, Habets, Emanuël A. P.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909321069068288
author Wechsler, Julian
Chetupalli, Srikanth Raj
Halimeh, Mhd Modar
Thiergart, Oliver
Habets, Emanuël A. P.
author_facet Wechsler, Julian
Chetupalli, Srikanth Raj
Halimeh, Mhd Modar
Thiergart, Oliver
Habets, Emanuël A. P.
contents Capturing audio signals with specific directivity patterns is essential in speech communication. This study presents a deep neural network (DNN)-based approach to directional filtering, alleviating the need for explicit signal models. More specifically, our proposed method uses a DNN to estimate a single-channel complex mask from the signals of a microphone array. This mask is then applied to a reference microphone to render a signal that exhibits a desired directivity pattern. We investigate the training dataset composition and its effect on the directivity realized by the DNN during inference. Using a relatively small DNN, the proposed method is found to approximate the desired directivity pattern closely. Additionally, it allows for the realization of higher-order directivity patterns using a small number of microphones, which is a difficult task for linear and parametric directional filtering.
format Preprint
id arxiv_https___arxiv_org_abs_2409_13502
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Neural Directional Filtering: Far-Field Directivity Control With a Small Microphone Array
Wechsler, Julian
Chetupalli, Srikanth Raj
Halimeh, Mhd Modar
Thiergart, Oliver
Habets, Emanuël A. P.
Audio and Speech Processing
Sound
Capturing audio signals with specific directivity patterns is essential in speech communication. This study presents a deep neural network (DNN)-based approach to directional filtering, alleviating the need for explicit signal models. More specifically, our proposed method uses a DNN to estimate a single-channel complex mask from the signals of a microphone array. This mask is then applied to a reference microphone to render a signal that exhibits a desired directivity pattern. We investigate the training dataset composition and its effect on the directivity realized by the DNN during inference. Using a relatively small DNN, the proposed method is found to approximate the desired directivity pattern closely. Additionally, it allows for the realization of higher-order directivity patterns using a small number of microphones, which is a difficult task for linear and parametric directional filtering.
title Neural Directional Filtering: Far-Field Directivity Control With a Small Microphone Array
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2409.13502