mlsanityData to 7 Oct 2026

Do End-to-End Speech Recognition Models Care About Context?

arXiv 2102.09928 ↗First posted 17 Feb 2021eess.AS cs.CL cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

The two most common paradigms for end-to-end speech recognition are connectionist temporal classification (CTC) and attention-based encoder-decoder (AED) models. It has been argued that the latter is better suited for learning an implicit language model. We test this hypothesis by measuring temporal context sensitivity and evaluate how the models perform when we constrain the amount of contextual information in the audio input. We find that the AED model is indeed more context sensitive, but that the gap can be closed by adding self-attention to the CTC model. Furthermore, the two models perform similarly when contextual information is constrained. Finally, in contrast to previous research, our results show that the CTC model is highly competitive on WSJ and LibriSpeech without the help of an external language model.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 23 of its 36 references that are in the map. First posted on arXiv 17 Feb 2021.

Builds on

23 of 36 references.

References · 23

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2016Densely Connected Convolutional Networkscited by 6,691
  3. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,530
  4. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  5. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,355
  6. 2015Listen, Attend and Spellcited by 542
  7. 2015Variational Dropout and the Local Reparameterization Trickcited by 489
  8. 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 449
  9. 2014Efficient Object Localization Using Convolutional Networkscited by 226
  10. 2016Understanding Neural Networks through Representation Erasurecited by 223
  11. 2015Visualizing and Understanding Neural Models in NLPcited by 199
  12. 2015End-to-End Attention-based Large Vocabulary Speech Recognitioncited by 148
11 more
  1. 2016Towards better decoding and language model integration in sequence to sequence modelscited by 100
  2. 2017Explaining Recurrent Neural Network Predictions in Sentiment Analysiscited by 87
  3. 2019Jasper: An End-to-End Convolutional Neural Acoustic Modelcited by 75
  4. 2019RWTH ASR Systems for LibriSpeech: Hybrid vs Attention -- w/o Data Augmentationcited by 63
  5. 2017Exploring Neural Transducers for End-to-End Speech Recognitioncited by 43
  6. 2016Very Deep Convolutional Networks for End-to-End Speech Recognitioncited by 34
  7. 2019Evaluating Recurrent Neural Network Explanationscited by 29
  8. 2018Fully Convolutional Speech Recognitioncited by 19
  9. 2018Optimal Completion Distillation for Sequence Learningcited by 14
  10. 2019On the Choice of Modeling Unit for Sequence-to-Sequence Speech Recognitioncited by 7
  11. 2016Latent Sequence Decompositionscited by 6

Built on by

Cited by year

Citing papers per year: 2021: 2; 2022: 0; 2023: 0; 2024: 0; 2025: 1.

3 papers.

  1. 2021Context-aware Adversarial Training for Name Regularity Bias in Named Entity Recognitioncited by 19
  2. 2021Speaker conditioning of acoustic models using affine transformation for multi-speaker speech recognitioncited by 2
  3. 2025Bridging ASR and LLMs for Dysarthric Speech Recognition: Benchmarking Self-Supervised and Generative Approachescited by 0