Gen-A: Generalizing Ambisonics Neural Encoding to Unseen Microphone Arrays

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Heikkinen, Mikko, Politis, Archontis, Drossos, Konstantinos, Virtanen, Tuomas
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915103061835776
author Heikkinen, Mikko
Politis, Archontis
Drossos, Konstantinos
Virtanen, Tuomas
author_facet Heikkinen, Mikko
Politis, Archontis
Drossos, Konstantinos
Virtanen, Tuomas
contents Using deep neural networks (DNNs) for encoding of microphone array (MA) signals to the Ambisonics spatial audio format can surpass certain limitations of established conventional methods, but existing DNN-based methods need to be trained separately for each MA. This paper proposes a DNN-based method for Ambisonics encoding that can generalize to arbitrary MA geometries unseen during training. The method takes as inputs the MA geometry and MA signals and uses a multi-level encoder consisting of separate paths for geometry and signal data, where geometry features inform the signal encoder at each level. The method is validated in simulated anechoic and reverberant conditions with one and two sources. The results indicate improvement over conventional encoding across the whole frequency range for dry scenes, while for reverberant scenes the improvement is frequency-dependent.
format Preprint
id arxiv_https___arxiv_org_abs_2501_08047
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Gen-A: Generalizing Ambisonics Neural Encoding to Unseen Microphone Arrays
Heikkinen, Mikko
Politis, Archontis
Drossos, Konstantinos
Virtanen, Tuomas
Audio and Speech Processing
Machine Learning
Sound
Using deep neural networks (DNNs) for encoding of microphone array (MA) signals to the Ambisonics spatial audio format can surpass certain limitations of established conventional methods, but existing DNN-based methods need to be trained separately for each MA. This paper proposes a DNN-based method for Ambisonics encoding that can generalize to arbitrary MA geometries unseen during training. The method takes as inputs the MA geometry and MA signals and uses a multi-level encoder consisting of separate paths for geometry and signal data, where geometry features inform the signal encoder at each level. The method is validated in simulated anechoic and reverberant conditions with one and two sources. The results indicate improvement over conventional encoding across the whole frequency range for dry scenes, while for reverberant scenes the improvement is frequency-dependent.
title Gen-A: Generalizing Ambisonics Neural Encoding to Unseen Microphone Arrays
topic Audio and Speech Processing
Machine Learning
Sound
url https://arxiv.org/abs/2501.08047