mlsanityData to 7 Oct 2026

Arabic Speech Recognition by End-to-End, Modular Systems and Human

arXiv 2101.08454 ↗First posted 21 Jan 2021eess.AS cs.CL cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

Recent advances in automatic speech recognition (ASR) have achieved accuracy levels comparable to human transcribers, which led researchers to debate if the machine has reached human performance. Previous work focused on the English language and modular hidden Markov model-deep neural network (HMM-DNN) systems. In this paper, we perform a comprehensive benchmarking for end-to-end transformer ASR, modular HMM-DNN ASR, and human speech recognition (HSR) on the Arabic language and its dialects. For the HSR, we evaluate linguist performance and lay-native speaker performance on a new dataset collected as a part of this study. For ASR the end-to-end work led to 12.5%, 27.5%, 33.8% WER; a new performance milestone for the MGB2, MGB3, and MGB5 challenges respectively. Our results suggest that human performance in the Arabic language is still considerably better than the machine with an absolute WER gap of 3.5% on average.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 21 of its 45 references that are in the map. First posted on arXiv 21 Jan 2021.

Builds on

21 of 45 references.

References · 21

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2016Layer Normalizationcited by 5,579
  4. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,355
  5. 2013Speech Recognition with Deep Recurrent Neural Networkscited by 666
  6. 2018Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidatescited by 632
  7. 2018ESPnet: End-to-End Speech Processing Toolkitcited by 554
  8. 2015Listen, Attend and Spellcited by 542
  9. 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 449
  10. 2015Attention-Based Models for Speech Recognitioncited by 400
  11. 2019A Comparative Study on Transformer vs RNN in Speech Applicationscited by 210
  12. 2019End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architecturescited by 120
9 more
  1. 2019Hierarchical Transformers for Long Document Classificationcited by 103
  2. 2019Transformer-based Acoustic Modeling for Hybrid Speech Recognitioncited by 86
  3. 2016Achieving Human Parity in Conversational Speech Recognitioncited by 63
  4. 2017Speech Recognition Challenge in the Wild: Arabic MGB-3cited by 60
  5. 2017English Conversational Telephone Speech Recognition by Humans and Machinescited by 40
  6. 2018Improved training of end-to-end attention models for speech recognitioncited by 35
  7. 2019Analyzing Phonetic and Graphemic Representations in End-to-End Automatic Speech Recognitioncited by 16
  8. 2017Comparing Human and Machine Errors in Conversational Speech Transcriptioncited by 7
  9. 2019DARTS: Dialectal Arabic Transcription Systemcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 1; 2023: 4; 2024: 1; 2025: 2; 2026: 1.

12 papers.

  1. 2021Towards One Model to Rule All: Multilingual Strategy for Dialectal Code-Switching Arabic ASRcited by 40
  2. 2021QASR: QCRI Aljazeera Speech Resource -- A Large Scale Annotated Arabic Speech Corpuscited by 38
  3. 2023N-Shot Benchmarking of Whisper on Diverse Arabic Speech Recognitioncited by 20
  4. 2024Casablanca: Data and Models for Multidialectal Arabic Speech Recognitioncited by 20
  5. 2022NatiQ: An End-to-end Text-to-Speech System for Arabiccited by 7
  6. 2023Unsupervised Data Selection for TTS: Using Arabic Broadcast News as a Case Studycited by 4
  7. 2021Dialectal Speech Recognition and Translation of Swiss German Speech to Standard German Text: Microsoft's Submission to SwissText 2021cited by 3
  8. 2023Quran Recitation Recognition using End-to-End Deep Learningcited by 3
  9. 2023VoxArabica: A Robust Dialect-Aware Arabic Speech Recognition Systemcited by 3
  10. 2025Advancing Arabic Speech Recognition Through Large-Scale Weakly Supervised Learningcited by 2
  11. 2026A Comparative Study of Pretrained Transformer Models for Quranic ASR: Speech Representations, Label Formats, and Dataset Compositioncited by 2
  12. 2025Open Automatic Speech Recognition Models for Classical and Modern Standard Arabiccited by 1