3M: Multi-loss, Multi-path and Multi-level Neural Networks for speech recognition

TL;DR

This paper introduces the 3M model for speech recognition, combining multi-loss, multi-path, and multi-level strategies to enhance performance, achieving significant CER reductions on public and large-scale datasets.

Contribution

The paper proposes a novel 3M architecture integrating multi-loss, multi-path, and multi-level techniques for improved speech recognition accuracy.

Findings

12.2%-17.6% relative CER improvement on WenetSpeech dataset

Superior performance over baseline Conformer on 150k hours dataset

Code is publicly available for reproducibility

Abstract

Recently, Conformer based CTC/AED model has become a mainstream architecture for ASR. In this paper, based on our prior work, we identify and integrate several approaches to achieve further improvements for ASR tasks, which we denote as multi-loss, multi-path and multi-level, summarized as "3M" model. Specifically, multi-loss refers to the joint CTC/AED loss and multi-path denotes the Mixture-of-Experts(MoE) architecture which can effectively increase the model capacity without remarkably increasing computation cost. Multi-level means that we introduce auxiliary loss at multiple level of a deep model to help training. We evaluate our proposed method on the public WenetSpeech dataset and experimental results show that the proposed method provides 12.2%-17.6% relative CER improvement over the baseline model trained by Wenet toolkit. On our large scale dataset of 150k hours corpus, the 3M model has also shown obvious superiority over the baseline Conformer model. Code is publicly available at https://github.com/tencent-ailab/3m-asr.