# Audio-attention discriminative language model for ASR rescoring

arXiv: https://arxiv.org/abs/1912.03363 · HTML: https://mlsanity.com/paper/1912.03363/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Dec 2019.

## Built on by (most cited first, 13 of 13)

- [Whispering LLaMA: A Cross-Modal Generative Error Correction Framework for Speech Recognition](https://mlsanity.com/paper/2310.06434/) (2023)
- [HyPoradise: An Open Baseline for Generative Speech Recognition with Large Language Models](https://mlsanity.com/paper/2309.15701/) (2023)
- [RescoreBERT: Discriminative Speech Recognition Rescoring with BERT](https://mlsanity.com/paper/2202.01094/) (2022)
- [Instant One-Shot Word-Learning for Context-Specific Neural Sequence-to-Sequence Speech Recognition](https://mlsanity.com/paper/2107.02268/) (2021)
- [ASR Rescoring and Confidence Estimation with ELECTRA](https://mlsanity.com/paper/2110.01857/) (2021)
- [Domain-aware Neural Language Models for Speech Recognition](https://mlsanity.com/paper/2101.03229/) (2021)
- [Low-rank Adaptation of Large Language Model Rescoring for Parameter-Efficient Speech Recognition](https://mlsanity.com/paper/2309.15223/) (2023)
- [Lattention: Lattice-attention in ASR rescoring](https://mlsanity.com/paper/2111.10157/) (2021)
- [Cross-Modal Transformer-Based Neural Correction Models for Automatic Speech Recognition](https://mlsanity.com/paper/2107.01569/) (2021)
- [Masked Audio Text Encoders are Effective Multi-Modal Rescorers](https://mlsanity.com/paper/2305.07677/) (2023)
- [Personalization for BERT-based Discriminative Speech Recognition Rescoring](https://mlsanity.com/paper/2307.06832/) (2023)
- [Speech Recognition Rescoring with Large Speech-Text Foundation Models](https://mlsanity.com/paper/2409.16654/) (2024)
- [Cross-utterance ASR Rescoring with Graph-based Label Propagation](https://mlsanity.com/paper/2303.15132/) (2023)

## Abstract (arXiv)

End-to-end approaches for automatic speech recognition (ASR) benefit from directly modeling the probability of the word sequence given the input audio stream in a single neural network. However, compared to conventional ASR systems, these models typically require more data to achieve comparable results. Well-known model adaptation techniques, to account for domain and style adaptation, are not easily applicable to end-to-end systems. Conventional HMM-based systems, on the other hand, have been optimized for various production environments and use cases. In this work, we propose to combine the benefits of end-to-end approaches with a conventional system using an attention-based discriminative language model that learns to rescore the output of a first-pass ASR system. We show that learning to rescore a list of potential ASR outputs is much simpler than learning to generate the hypothesis. The proposed model results in 8% improvement in word error rate even when the amount of training data is a fraction of data used for training the first-pass system.