Neural Ambisonic Encoding For Multi-Speaker Scenarios Using A Circular Microphone Array

TL;DR

This paper introduces a deep learning approach for encoding circular microphone array signals into second-order Ambisonics in multi-speaker scenarios, improving spatial audio quality and source localization over traditional methods.

Contribution

A novel two-stage neural network architecture with a spatial power map loss and channel permutation technique for Ambisonic encoding from circular arrays.

Findings

Outperforms traditional signal processing methods in spatial quality.

Achieves higher source localization accuracy.

Provides better timbral quality in simulated tests.

Abstract

Spatial audio formats like Ambisonics are playback device layout-agnostic and well-suited for applications such as teleconferencing and virtual reality. Conventional Ambisonic encoding methods often rely on spherical microphone arrays for efficient sound field capture, which limits their flexibility in practical scenarios. We propose a deep learning (DL)-based approach, leveraging a two-stage network architecture for encoding circular microphone array signals into second-order Ambisonics (SOA) in multi-speaker environments. In addition, we introduce: (i) a novel loss function based on spatial power maps to regularize inter-channel correlations of the Ambisonic signals, and (ii) a channel permutation technique to resolve the ambiguity of encoding vertical information using a horizontal circular array. Evaluation on simulated speech and noise datasets shows that our approach consistently outperforms traditional signal processing (SP) and DL-based methods, providing significantly better timbral and spatial quality and higher source localization accuracy. Binaural audio demos with visualizations are available at https://bridgoon97.github.io/NeuralAmbisonicEncoding/.