Efficient learning-based sound propagation for virtual and real-world audio processing applications

TL;DR

This paper presents a fast, learning-based method for generating and estimating room impulse responses (RIRs) that enhances audio processing tasks like speech recognition and synthesis, outperforming traditional simulators.

Contribution

It introduces a novel learning-based RIR generator, an RIR estimator from reverberant speech and visual cues, and IR-GAN for augmenting RIRs, advancing acoustic simulation and estimation techniques.

Findings

Learning-based RIR generator is two orders of magnitude faster than ray-tracing.

Estimated RIRs improve far-field ASR word error rate by 6.9%.

IR-GAN outperforms ray-tracing in ASR benchmark by 8.95%.

Abstract

Sound propagation is the process by which sound energy travels through a medium, such as air, to the surrounding environment as sound waves. The room impulse response (RIR) describes this process and is influenced by the positions of the source and listener, the room's geometry, and its materials. Physics-based acoustic simulators have been used for decades to compute accurate RIRs for specific acoustic environments. However, we have encountered limitations with existing acoustic simulators. To address these limitations, we propose three novel solutions. First, we introduce a learning-based RIR generator that is two orders of magnitude faster than an interactive ray-tracing simulator. Our approach can be trained to input both statistical and traditional parameters directly, and it can generate both monaural and binaural RIRs for both reconstructed and synthetic 3D scenes. Our generated RIRs outperform interactive ray-tracing simulators in speech-processing applications, including ASR, Speech Enhancement, and Speech Separation. Secondly, we propose estimating RIRs from reverberant speech signals and visual cues without a 3D representation of the environment. By estimating RIRs from reverberant speech, we can augment training data to match test data, improving the word error rate of the ASR system. Our estimated RIRs achieve a 6.9% improvement over previous learning-based RIR estimators in far-field ASR tasks. We demonstrate that our audio-visual RIR estimator aids tasks like visual acoustic matching, novel-view acoustic synthesis, and voice dubbing, validated through perceptual evaluation. Finally, we introduce IR-GAN to augment accurate RIRs using real RIRs. IR-GAN parametrically controls acoustic parameters learned from real RIRs to generate new RIRs that imitate different acoustic environments, outperforming Ray-tracing simulators on the far-field ASR benchmark by 8.95%.