mlsanityData to 7 Oct 2026

Cascade versus Direct Speech Translation: Do the Differences Still Make a Difference?

arXiv 2106.01045 ↗First posted 2 Jun 2021cs.CLcited by 54

Not on the map. Why?

Abstract (arXiv)

Five years after the first published proofs of concept, direct approaches to speech translation (ST) are now competing with traditional cascade solutions. In light of this steady progress, can we claim that the performance gap between the two is closed? Starting from this question, we present a systematic comparison between state-of-the-art systems representative of the two paradigms. Focusing on three language directions (English-German/Italian/Spanish), we conduct automatic and manual evaluations, exploiting high-quality professional post-edits and annotations. Our multi-faceted analysis on one of the few publicly available ST benchmarks attests for the first time that: i) the gap between the two paradigms is now closed, and ii) the subtle differences observed in their behavior are not sufficient for humans neither to distinguish them nor to prefer one over the other.

Key facts. Cited by 54 papers in the map (data through 7 Oct 2026). Builds on 31 of its 72 references that are in the map. First posted on arXiv 2 Jun 2021.

Builds on

31 of 72 references.

References · 31

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  3. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  4. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,898
  5. 2018A Call for Clarity in Reporting BLEU Scorescited by 1,790
  6. 2019fairseq: A Fast, Extensible Toolkit for Sequence Modelingcited by 1,486
  7. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,355
  8. 2018TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptationcited by 224
  9. 2018How2: A Large-scale Dataset for Multimodal Language Understandingcited by 137
  10. 2017Sequence-to-Sequence Models Can Directly Translate Foreign Speechcited by 121
  11. 2019Europarl-ST: A Multilingual Corpus For Speech Translation Of Parliamentary Debatescited by 114
  12. 2016Listen and Translate: A Proof of Concept for End-to-End Speech-to-Text Translationcited by 113
19 more
  1. 2020ESPnet-ST: All-in-One Speech Translation Toolkitcited by 82
  2. 2020Speech Translation and the End-to-End Promise: Taking Stock of Where We Arecited by 77
  3. 2018Pre-training on high-resource speech recognition improves low-resource speech-to-text translationcited by 70
  4. 2018End-to-End Automatic Speech Translation of Audiobookscited by 65
  5. 2019End-to-End Speech Translation with Knowledge Distillationcited by 58
  6. 2018Leveraging Weakly Supervised Data to Improve End-to-End Speech-to-Text Translationcited by 57
  7. 2020Bridging the Modality Gap for Speech-to-Text Translationcited by 55
  8. 2018Tied Multitask Learning for Neural Speech Translationcited by 54
  9. 2019A Comparative Study on End-to-end Speech to Text Translationcited by 40
  10. 2020End-to-End Speech-Translation with Knowledge Distillation: FBK@IWSLT2020cited by 31
  11. 2018Self-Attentional Acoustic Modelscited by 29
  12. 2019Improving sequence-to-sequence speech recognition training with on-the-fly data augmentationcited by 29
  13. 2019On Using SpecAugment for End-to-End Speech Translationcited by 19
  14. 2019One-To-Many Multilingual End-to-end Speech Translationcited by 17
  15. 2018XNMT: The eXtensible Neural Machine Translation Toolkitcited by 14
  16. 2019Robust Neural Machine Translation for Clean and Noisy Speech Transcriptscited by 13
  17. 2019Self-Attentional Models for Lattice Inputscited by 12
  18. 2018Quality expectations of machine translationcited by 9
  19. 2019Breaking the Data Barrier: Towards Robust Speech Translation via Adversarial Stability Trainingcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 15; 2023: 14; 2024: 10; 2025: 7; 2026: 4.

54 papers.

Sort:

  1. 2023Hallucinations in Large Multilingual Translation Modelscited by 126
  2. 2022Cross-modal Contrastive Learning for Speech Translationcited by 61
  3. 2022SHAS: Approaching optimal Segmentation for End-to-End Speech Translationcited by 26
  4. 2023Pre-training for Speech Translation: CTC Meets Optimal Transportcited by 25
  5. 2022Under the Morphosyntactic Lens: A Multifaceted Evaluation of Gender Bias in Speech Translationcited by 23
  6. 2024Speech Translation with Speech Foundation Models and Large Language Models: What is There and What is Missing?cited by 23
  7. 2024How "Real" is Your Real-Time Simultaneous Speech-to-Text Translation System?cited by 17
  8. 2023Understanding and Bridging the Modality Gap for Speech Translationcited by 15
  9. 2023Recent Advances in Direct Speech-to-text Translationcited by 15
  10. 2024Pushing the Limits of Zero-shot End-to-End Speech Translationcited by 14
  11. 2022Non-Parametric Domain Adaptation for End-to-End Speech Translationcited by 12
  12. 2022Simple and Effective Unsupervised Speech Translationcited by 12
38 more
  1. 2024LLaST: Improved End-to-end Speech Translation System Leveraged by Large Language Modelscited by 12
  2. 2023ESPnet-ST-v2: Multipurpose Spoken Language Translation Toolkitcited by 10
  3. 2021Non-autoregressive End-to-end Speech Translation with Parallel Autoregressive Rescoringcited by 9
  4. 2023End-to-End Speech-to-Text Translation: A Surveycited by 9
  5. 2022Who Are We Talking About? Handling Person Names in Speech Translationcited by 8
  6. 2022Direct Speech Translation for Automatic Subtitlingcited by 7
  7. 2022ON-TRAC Consortium Systems for the IWSLT 2022 Dialect and Low-resource Speech Translation Taskscited by 6
  8. 2022SegAugment: Maximizing the Utility of Speech Translation Data with Segmentation-based Augmentationscited by 6
  9. 2023Direct Models for Simultaneous Translation and Automatic Subtitling: FBK@IWSLT2023cited by 6
  10. 2025Hearing to Translate: The Effectiveness of Speech Modality Integration into LLMscited by 6
  11. 2021Efficient Transformer for Direct Speech Translationcited by 5
  12. 2022Joint Speech Translation and Named Entity Recognitioncited by 5
  13. 2023Improving Cascaded Unsupervised Speech Translation with Denoising Back-translationcited by 5
  14. 2024Prosody in Cascade and Direct Speech-to-Text Translation: a case study on Korean Wh-Phrasescited by 5
  15. 2024Evaluating the IWSLT2023 Speech Translation Tasks: Human Annotations, Automatic Metrics, and Segmentationcited by 5
  16. 2024Label-Synchronous Neural Transducer for E2E Simultaneous Speech Translationcited by 5
  17. 2022Named Entity Detection and Injection for Direct Speech Translationcited by 4
  18. 2023Towards Speech Dialogue Translation Mediating Speakers of Different Languagescited by 4
  19. 2024Towards Building Large Scale Datasets and State-of-the-Art Automatic Speech Translation Systems for 14 Indian Languagescited by 4
  20. 2023Strategies for improving low resource speech to text translation relying on pre-trained ASR modelscited by 3
  21. 2023Long-form Simultaneous Speech Translation: Thesis Proposalcited by 3
  22. 2025Different Speech Translation Models Encode and Translate Speaker Gender Differentlycited by 3
  23. 2025How to Evaluate Speech Translation with Source-Aware Neural MT Metricscited by 3
  24. 2021Visualization: the missing factor in Simultaneous Speech Translationcited by 2
  25. 2022Efficient Speech Translation with Dynamic Latent Perceiverscited by 2
  26. 2023Improving Autoregressive NLP Tasks via Modular Linearized Attentioncited by 2
  27. 2023SpeechAlign: a Framework for Speech Translation Alignment Evaluationcited by 2
  28. 2024A Case Study on Filtering for End-to-End Speech Translationcited by 2
  29. 2022Multiformer: A Head-Configurable Transformer-Based Model for Direct Speech Translationcited by 1
  30. 2022Efficient Speech Translation with Pre-trained Modelscited by 1
  31. 2023Speech Translation with Foundation Models and Optimal Transport: UPC at IWSLT23cited by 1
  32. 2024CrossVoice: Crosslingual Prosody Preserving Cascade-S2ST using Transfer Learningcited by 1
  33. 2025IIITH-BUT system for IWSLT 2025 low-resource Bhojpuri to Hindi speech translationcited by 1
  34. 2021The HW-TSC's Offline Speech Translation Systems for IWSLT 2021 Evaluationcited by 0
  35. 2022JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMTcited by 0
  36. 2025Speech to Speech Translation with Translatotron: A State of the Art Reviewcited by 0
  37. 2025Speech-to-Speech Translation Pipelines for Conversations in Low-Resource Languagescited by 0
  38. 2025Optimal Multi-Task Learning at Regularization Horizon for Speech Translation Taskcited by 0

50 most cited of 54 All 54 (JSON)