# Speech-text based multi-modal training with bidirectional attention for improved speech recognition

arXiv: https://arxiv.org/abs/2211.00325 · HTML: https://mlsanity.com/paper/2211.00325/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 39 references that are in the map.
- First posted on arXiv 1 Nov 2022.

## Builds on (most cited first, 30 of 33)

- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://mlsanity.com/paper/2202.03555/) (2022)
- [ESPnet: End-to-End Speech Processing Toolkit](https://mlsanity.com/paper/1804.00015/) (2018)
- [Listen, Attend and Spell](https://mlsanity.com/paper/1508.01211/) (2015)
- [vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations](https://mlsanity.com/paper/1910.05453/) (2019)
- [Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss](https://mlsanity.com/paper/2002.02562/) (2020)
- [Streaming End-to-end Speech Recognition For Mobile Devices](https://mlsanity.com/paper/1811.06621/) (2018)
- [Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders](https://mlsanity.com/paper/1910.12638/) (2019)
- [State-of-the-art Speech Recognition With Sequence-to-Sequence Models](https://mlsanity.com/paper/1712.01769/) (2017)
- [An Unsupervised Autoregressive Model for Speech Representation Learning](https://mlsanity.com/paper/1904.03240/) (2019)
- [SpeechT5: Unified-Modal Encoder-Decoder Pre-Training for Spoken Language Processing](https://mlsanity.com/paper/2110.07205/) (2021)
- [Towards better decoding and language model integration in sequence to sequence models](https://mlsanity.com/paper/1612.02695/) (2016)
- [A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency](https://mlsanity.com/paper/2003.12710/) (2020)
- [Cold Fusion: Training Seq2Seq Models Together with Language Models](https://mlsanity.com/paper/1708.06426/) (2017)
- [MAESTRO: Matched Speech Text Representations through Modality Matching](https://mlsanity.com/paper/2204.03409/) (2022)
- [CIF: Continuous Integrate-and-Fire for End-to-End Speech Recognition](https://mlsanity.com/paper/1905.11235/) (2019)
- [Deep Contextualized Acoustic Representations For Semi-Supervised Speech Recognition](https://mlsanity.com/paper/1912.01679/) (2019)
- [A General Multi-Task Learning Framework to Leverage Text Data for Speech to Text Tasks](https://mlsanity.com/paper/2010.11338/) (2020)
- [Leveraging Unpaired Text Data for Training End-to-End Speech-to-Intent Systems](https://mlsanity.com/paper/2010.04284/) (2020)
- [Cycle-consistency training for end-to-end speech recognition](https://mlsanity.com/paper/1811.01690/) (2018)
- [Noise-robust Speech Recognition with 10 Minutes Unparalleled In-domain Data](https://mlsanity.com/paper/2203.15321/) (2022)
- [Multi-Modal Data Augmentation for End-to-End ASR](https://mlsanity.com/paper/1803.10299/) (2018)
- [Interactive Feature Fusion for End-to-End Noise-Robust Speech Recognition](https://mlsanity.com/paper/2110.05267/) (2021)
- [Wav-BERT: Cooperative Acoustic and Linguistic Representation Learning for Low-Resource Speech Recognition](https://mlsanity.com/paper/2109.09161/) (2021)
- [Improving CTC-based speech recognition via knowledge transferring from pre-trained language models](https://mlsanity.com/paper/2203.03582/) (2022)
- [Optimizing Alignment of Speech and Language Latent Spaces for End-to-End Speech Recognition and Understanding](https://mlsanity.com/paper/2110.12138/) (2021)
- [NeuFA: Neural Network Based End-to-End Forced Alignment with Bidirectional Attention Mechanism](https://mlsanity.com/paper/2203.16838/) (2022)

## Built on by (most cited first, 4 of 4)

- [Unifying Speech Enhancement and Separation with Gradient Modulation for End-to-End Noise-Robust Speech Separation](https://mlsanity.com/paper/2302.11131/) (2023)
- [Gradient Remedy for Multi-Task Learning in End-to-End Noise-Robust Speech Recognition](https://mlsanity.com/paper/2302.11362/) (2023)
- [CLIP-Decoder : ZeroShot Multilabel Classification using Multimodal CLIP Aligned Representation](https://mlsanity.com/paper/2406.14830/) (2024)
- [BLR-MoE: Boosted Language-Routing Mixture of Experts for Domain-Robust Multilingual E2E ASR](https://mlsanity.com/paper/2501.12602/) (2025)

## Abstract (arXiv)

To let the state-of-the-art end-to-end ASR model enjoy data efficiency, as well as much more unpaired text data by multi-modal training, one needs to address two problems: 1) the synchronicity of feature sampling rates between speech and language (aka text data); 2) the homogeneity of the learned representations from two encoders. In this paper we propose to employ a novel bidirectional attention mechanism (BiAM) to jointly learn both ASR encoder (bottom layers) and text encoder with a multi-modal learning method. The BiAM is to facilitate feature sampling rate exchange, realizing the quality of the transformed features for the one kind to be measured in another space, with diversified objective functions. As a result, the speech representations are enriched with more linguistic information, while the representations generated by the text encoder are more similar to corresponding speech ones, and therefore the shared ASR models are more amenable for unpaired text data pretraining. To validate the efficacy of the proposed method, we perform two categories of experiments with or without extra unpaired text data. Experimental results on Librispeech corpus show it can achieve up to 6.15% word error rate reduction (WERR) with only paired data learning, while 9.23% WERR when more unpaired text data is employed.