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

arXiv: https://arxiv.org/abs/2204.03178 · HTML: https://mlsanity.com/paper/2204.03178/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal MoE models, one of 12 directions in Mixture-of-experts.
- Builds on 12 of its 20 references that are in the map.
- First posted on arXiv 7 Apr 2022.

## Placed in

- Mixture-of-experts / [Multimodal MoE models](https://mlsanity.com/moe/multimodal-moe-models/)

## Builds on (most cited first, 12 of 12)

- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [ESPnet: End-to-End Speech Processing Toolkit](https://mlsanity.com/paper/1804.00015/) (2018)
- [WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition](https://mlsanity.com/paper/2110.03370/) (2021)
- [SpeechStew: Simply Mix All Available Speech Recognition Data to Train One Large Neural Network](https://mlsanity.com/paper/2104.02133/) (2021)
- [Hard Mixtures of Experts for Large Scale Weakly Supervised Vision](https://mlsanity.com/paper/1704.06363/) (2017)
- [Network of Experts for Large-Scale Image Categorization](https://mlsanity.com/paper/1604.06119/) (2016)
- [SpeechMoE: Scaling to Large Acoustic Models with Dynamic Routing Mixture of Experts](https://mlsanity.com/paper/2105.03036/) (2021)
- [Advances in Joint CTC-Attention based End-to-End Speech Recognition with a Deep CNN Encoder and RNN-LM](https://mlsanity.com/paper/1706.02737/) (2017)
- [SpeechMoE2: Mixture-of-Experts Model with Improved Routing](https://mlsanity.com/paper/2111.11831/) (2021)

## Built on by (most cited first, 5 of 5)

- [Zipformer: A faster and better encoder for automatic speech recognition](https://mlsanity.com/paper/2310.11230/) (2023)
- [eMoE: Task-aware Memory Efficient Mixture-of-Experts-Based (MoE) Model Inference](https://mlsanity.com/paper/2503.06823/) (2025)
- [CIF-T: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition](https://mlsanity.com/paper/2307.14132/) (2023)
- [U2++ MoE: Scaling 4.7x parameters with minimal impact on RTF](https://mlsanity.com/paper/2404.16407/) (2024)
- [UME: Upcycling Mixture-of-Experts for Scalable and Efficient Automatic Speech Recognition](https://mlsanity.com/paper/2412.17507/) (2024)

## Abstract (arXiv)

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.