mlsanityData to 14 Oct 2026

Are E2E ASR models ready for an industrial usage?

arXiv 2112.12572 ↗First posted 9 Dec 2021eess.AS cs.AI cs.CLcited by 3

Not on the map. Why?

Abstract (arXiv)

The Automated Speech Recognition (ASR) community experiences a major turning point with the rise of the fully-neural (End-to-End, E2E) approaches. At the same time, the conventional hybrid model remains the standard choice for the practical usage of ASR. According to previous studies, the adoption of E2E ASR in real-world applications was hindered by two main limitations: their ability to generalize on unseen domains and their high operational cost. In this paper, we investigate both above-mentioned drawbacks by performing a comprehensive multi-domain benchmark of several contemporary E2E models and a hybrid baseline. Our experiments demonstrate that E2E models are viable alternatives for the hybrid approach, and even outperform the baseline both in accuracy and in operational efficiency. As a result, our study shows that the generalization and complexity issues are no longer the major obstacle for industrial integration, and draws the community's attention to other potential limitations of the E2E approaches in some specific use-cases.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 21 of its 31 references that are in the map. First posted on arXiv 9 Dec 2021.

Builds on

21 of 31 references.

References · 21

  1. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
  2. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  3. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,380
  4. 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 451
  5. 2021SpeechBrain: A General-Purpose Speech Toolkitcited by 406
  6. 2019NeMo: a toolkit for building AI applications using Neural Modulescited by 235
  7. 2018TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptationcited by 226
  8. 2020ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Contextcited by 97
  9. 2020Iterative Pseudo-Labeling for Speech Recognitioncited by 83
  10. 2020Rethinking Evaluation in ASR: Are Our Models Robust Enough?cited by 73
  11. 2020FastEmit: Low-latency Streaming ASR with Sequence-level Emission Regularizationcited by 69
  12. 2021LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speechcited by 47
9 more
  1. 2018wav2letter++: The Fastest Open-source Speech Recognition Systemcited by 42
  2. 2021Efficient conformer: Progressive downsampling and grouped attention for automatic speech recognitioncited by 42
  3. 2018Toward domain-invariant speech recognition via large scale trainingcited by 35
  4. 2021Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognitioncited by 34
  5. 2021Large-Scale Pre-Training of End-to-End Multi-Talker ASR for Meeting Transcription with Single Distant Microphonecited by 33
  6. 2021NeMo Inverse Text Normalization: From Development To Productioncited by 29
  7. 2020WER we are and WER we think we arecited by 19
  8. 2021Efficient conformer-based speech recognition with linear attentioncited by 13
  9. 2021Efficient Conformer with Prob-Sparse Attention Mechanism for End-to-EndSpeech Recognitioncited by 9

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 1.

3 papers.

  1. 2023Improving Speech Recognition for African American English With Audio Classificationcited by 4
  2. 2024Deep CLAS: Deep Contextual Listen, Attend and Spellcited by 2
  3. 2022Goodness of Pronunciation Pipelines for OOV Problemcited by 0