Multi-scale Octave Convolutions for Robust Speech Recognition

TL;DR

This paper introduces a multi-scale octave convolution layer for speech recognition, enhancing robustness to noise and computational efficiency by decomposing features into multiple resolutions, inspired by vision models.

Contribution

It extends octave convolutions to multiple resolutions in speech models, demonstrating improved noise robustness and efficiency in CNN-based speech recognition systems.

Findings

Reduces WER by up to 6.6% on Aurora-4

Improves robustness to noisy conditions

Enhances computational efficiency of CNN models

Abstract

We propose a multi-scale octave convolution layer to learn robust speech representations efficiently. Octave convolutions were introduced by Chen et al [1] in the computer vision field to reduce the spatial redundancy of the feature maps by decomposing the output of a convolutional layer into feature maps at two different spatial resolutions, one octave apart. This approach improved the efficiency as well as the accuracy of the CNN models. The accuracy gain was attributed to the enlargement of the receptive field in the original input space. We argue that octave convolutions likewise improve the robustness of learned representations due to the use of average pooling in the lower resolution group, acting as a low-pass filter. We test this hypothesis by evaluating on two noisy speech corpora - Aurora-4 and AMI. We extend the octave convolution concept to multiple resolution groups and multiple octaves. To evaluate the robustness of the inferred representations, we report the similarity between clean and noisy encodings using an affine projection loss as a proxy robustness measure. The results show that proposed method reduces the WER by up to 6.6% relative for Aurora-4 and 3.6% for AMI, while improving the computational efficiency of the CNN acoustic models.