Transformer-based Acoustic Modeling for Hybrid Speech Recognition

TL;DR

This paper introduces transformer-based acoustic models for hybrid speech recognition, demonstrating significant improvements over previous methods and enabling streaming applications with limited right context.

Contribution

It presents novel transformer modeling choices, including positional embeddings and an iterated loss, achieving state-of-the-art results on Librispeech.

Findings

Outperforms previous hybrid models by 19-26% on Librispeech

Enables streaming with limited right context

Achieves state-of-the-art results with neural LM rescoring

Abstract

We propose and evaluate transformer-based acoustic models (AMs) for hybrid speech recognition. Several modeling choices are discussed in this work, including various positional embedding methods and an iterated loss to enable training deep transformers. We also present a preliminary study of using limited right context in transformer models, which makes it possible for streaming applications. We demonstrate that on the widely used Librispeech benchmark, our transformer-based AM outperforms the best published hybrid result by 19% to 26% relative when the standard n-gram language model (LM) is used. Combined with neural network LM for rescoring, our proposed approach achieves state-of-the-art results on Librispeech. Our findings are also confirmed on a much larger internal dataset.