Attention-based ASR with Lightweight and Dynamic Convolutions

TL;DR

This paper introduces a novel end-to-end speech recognition architecture using lightweight and dynamic convolutions as an efficient alternative to self-attention, achieving competitive results with state-of-the-art models.

Contribution

It proposes a new convolution-based architecture for E2E ASR that reduces computational complexity and enhances performance, including joint training and frequency axis convolution.

Findings

Achieves better performance than RNN-based models.

Competitive results with Transformer on various benchmarks.

Effective in noisy and reverberant environments.

Abstract

End-to-end (E2E) automatic speech recognition (ASR) with sequence-to-sequence models has gained attention because of its simple model training compared with conventional hidden Markov model based ASR. Recently, several studies report the state-of-the-art E2E ASR results obtained by Transformer. Compared to recurrent neural network (RNN) based E2E models, training of Transformer is more efficient and also achieves better performance on various tasks. However, self-attention used in Transformer requires computation quadratic in its input length. In this paper, we propose to apply lightweight and dynamic convolution to E2E ASR as an alternative architecture to the self-attention to make the computational order linear. We also propose joint training with connectionist temporal classification, convolution on the frequency axis, and combination with self-attention. With these techniques, the proposed architectures achieve better performance than RNN-based E2E model and performance competitive to state-of-the-art Transformer on various ASR benchmarks including noisy/reverberant tasks.