Multi-Channel Transformer Transducer for Speech Recognition

TL;DR

This paper introduces the Multi-Channel Transformer Transducer (MCTT), a low-complexity, end-to-end model for multi-channel speech recognition that achieves significant accuracy improvements and faster inference suitable for on-device streaming applications.

Contribution

The paper proposes MCTT, a novel multi-channel transformer transducer model with reduced computational cost and latency, enabling effective on-device streaming speech recognition.

Findings

MCTT achieves up to 6.01% relative WER improvement over stagewise models.

MCTT outperforms multi-channel transformer by 11.62% WERR.

MCTT is 15.8 times faster in inference speed.

Abstract

Multi-channel inputs offer several advantages over single-channel, to improve the robustness of on-device speech recognition systems. Recent work on multi-channel transformer, has proposed a way to incorporate such inputs into end-to-end ASR for improved accuracy. However, this approach is characterized by a high computational complexity, which prevents it from being deployed in on-device systems. In this paper, we present a novel speech recognition model, Multi-Channel Transformer Transducer (MCTT), which features end-to-end multi-channel training, low computation cost, and low latency so that it is suitable for streaming decoding in on-device speech recognition. In a far-field in-house dataset, our MCTT outperforms stagewise multi-channel models with transformer-transducer up to 6.01% relative WER improvement (WERR). In addition, MCTT outperforms the multi-channel transformer up to 11.62% WERR, and is 15.8 times faster in terms of inference speed. We further show that we can improve the computational cost of MCTT by constraining the future and previous context in attention computations.