Invariant Representations for Noisy Speech Recognition

TL;DR

This paper proposes a neural network architecture that learns invariant feature representations to improve noise robustness in speech recognition, outperforming standard methods especially with limited noise conditions seen during training.

Contribution

It introduces a novel adversarial training approach inspired by GANs and domain adaptation to produce noise-invariant features for ASR systems.

Findings

Improved generalization to unseen noise conditions.

Better robustness compared to standard multi-condition training.

Effective with limited noise categories during training.

Abstract

Modern automatic speech recognition (ASR) systems need to be robust under acoustic variability arising from environmental, speaker, channel, and recording conditions. Ensuring such robustness to variability is a challenge in modern day neural network-based ASR systems, especially when all types of variability are not seen during training. We attempt to address this problem by encouraging the neural network acoustic model to learn invariant feature representations. We use ideas from recent research on image generation using Generative Adversarial Networks and domain adaptation ideas extending adversarial gradient-based training. A recent work from Ganin et al. proposes to use adversarial training for image domain adaptation by using an intermediate representation from the main target classification network to deteriorate the domain classifier performance through a separate neural network. Our work focuses on investigating neural architectures which produce representations invariant to noise conditions for ASR. We evaluate the proposed architecture on the Aurora-4 task, a popular benchmark for noise robust ASR. We show that our method generalizes better than the standard multi-condition training especially when only a few noise categories are seen during training.