mlsanityData to 14 Oct 2026

On Language Model Integration for RNN Transducer based Speech Recognition

arXiv 2110.06841 ↗First posted 13 Oct 2021cs.CL eess.AScited by 17

Not on the map. Why?

Abstract (arXiv)

The mismatch between an external language model (LM) and the implicitly learned internal LM (ILM) of RNN-Transducer (RNN-T) can limit the performance of LM integration such as simple shallow fusion. A Bayesian interpretation suggests to remove this sequence prior as ILM correction. In this work, we study various ILM correction-based LM integration methods formulated in a common RNN-T framework. We provide a decoding interpretation on two major reasons for performance improvement with ILM correction, which is further experimentally verified with detailed analysis. We also propose an exact-ILM training framework by extending the proof given in the hybrid autoregressive transducer, which enables a theoretical justification for other ILM approaches. Systematic comparison is conducted for both in-domain and cross-domain evaluation on the Librispeech and TED-LIUM Release 2 corpora, respectively. Our proposed exact-ILM training can further improve the best ILM method.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Builds on 22 of its 34 references that are in the map. First posted on arXiv 13 Oct 2021.

Builds on

22 of 34 references.

References · 22

  1. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  2. 2015Listen, Attend and Spellcited by 543
  3. 2014Deep Speech: Scaling up end-to-end speech recognitioncited by 354
  4. 2015End-to-End Attention-based Large Vocabulary Speech Recognitioncited by 151
  5. 2015On Using Monolingual Corpora in Neural Machine Translationcited by 150
  6. 2020Hybrid Autoregressive Transducer (hat)cited by 108
  7. 2019Language Modeling with Deep Transformerscited by 96
  8. 2020Internal Language Model Estimation for Domain-Adaptive End-to-End Speech Recognitioncited by 80
  9. 2020A Density Ratio Approach to Language Model Fusion in End-To-End Automatic Speech Recognitioncited by 73
  10. 2021Advancing RNN Transducer Technology for Speech Recognitioncited by 44
  11. 2021Internal Language Model Training for Domain-Adaptive End-to-End Speech Recognitioncited by 37
  12. 2020Single headed attention based sequence-to-sequence model for state-of-the-art results on Switchboardcited by 36
10 more
  1. 2021Librispeech Transducer Model with Internal Language Model Prior Correctioncited by 26
  2. 2021Investigating Methods to Improve Language Model Integration for Attention-based Encoder-Decoder ASR Modelscited by 25
  3. 2020Phoneme Based Neural Transducer for Large Vocabulary Speech Recognitioncited by 20
  4. 2021Fast Text-Only Domain Adaptation of RNN-Transducer Prediction Networkcited by 19
  5. 2021Internal Language Model Adaptation with Text-Only Data for End-to-End Speech Recognitioncited by 17
  6. 2020On Minimum Word Error Rate Training of the Hybrid Autoregressive Transducercited by 11
  7. 2021Acoustic Data-Driven Subword Modeling for End-to-End Speech Recognitioncited by 11
  8. 2020Early Stage LM Integration Using Local and Global Log-Linear Combinationcited by 9
  9. 2021On Architectures and Training for Raw Waveform Feature Extraction in ASRcited by 8
  10. 2021Comparing the Benefit of Synthetic Training Data for Various Automatic Speech Recognition Architecturescited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 4; 2023: 10; 2024: 0; 2025: 2; 2026: 1.

17 papers.

Sort:

  1. 2023End-to-End Speech Recognition: A Surveycited by 129
  2. 2022Towards Zero-Shot Code-Switched Speech Recognitioncited by 15
  3. 2022Efficient Training of Neural Transducer for Speech Recognitioncited by 14
  4. 2023CB-Conformer: Contextual biasing Conformer for biased word recognitioncited by 11
  5. 2023ed-cec: improving rare word recognition using asr postprocessing based on error detection and context-aware error correctioncited by 7
  6. 2025Delayed Fusion: Integrating Large Language Models into First-Pass Decoding in End-to-end Speech Recognitioncited by 5
  7. 2022Lattice-Free Sequence Discriminative Training for Phoneme-Based Neural Transducerscited by 3
  8. 2023RASR2: The RWTH ASR Toolkit for Generic Sequence-to-sequence Speech Recognitioncited by 3
  9. 2023Decoupled Structure for Improved Adaptability of End-to-End Modelscited by 3
  10. 2022An Empirical Study of Language Model Integration for Transducer based Speech Recognitioncited by 2
  11. 2023External Language Model Integration for Factorized Neural Transducerscited by 2
  12. 2023Improved Factorized Neural Transducer Model For text-only Domain Adaptationcited by 2
5 more
  1. 2025Label-Context-Dependent Internal Language Model Estimation for CTCcited by 2
  2. 2023Mask The Bias: Improving Domain-Adaptive Generalization of CTC-based ASR with Internal Language Model Estimationcited by 1
  3. 2023On the Relation between Internal Language Model and Sequence Discriminative Training for Neural Transducerscited by 1
  4. 2023Investigating the Effect of Language Models in Sequence Discriminative Training for Neural Transducerscited by 0
  5. 2026Revisiting the Relation Between Language Model Perplexity and ASR Word Error Rate for Modern End-to-End Speech Recognitioncited by 0