mlsanityData to 19 Oct 2026

Speech translation

365 papers0 this weekHigh confidence in this grouping: the papers form one clear topicHigh confidence in this grouping: the papers form one clear topic · About

Keywords: translation, speech translation, speech, simultaneous, language

Key facts. Speech translation: 365 papers on arXiv since 2021 (data through 19 Oct 2026). 70 of its 365 mapped papers (19%) cite “Direct speech-to-speech translation with discrete units” (2021, arXiv 2107.05604). 4 new papers 2 Oct–19 Oct.

What’s new

2 Oct – 19 Oct 2026

  1. Earlier · 4
  2. Cross-Lingual TTS for Indian Languages: A Benchmark and Inference-Time Duration Estimation Strategies also fits: Multilingual and low-resource ASR
  3. Dialect-Robust Speech Language Models with Synthetic Pseudo-Dialect Augmentation
  4. Quality-Aware Self-Correcting Speech Translation on an Edge Device
  5. Learning When to Commit from Partial Speech for End-to-End Simultaneous Speech Translation

Foundations

Most cited by papers here

  1. Direct speech-to-speech translation with discrete units2021cited by 70 here
  2. Translatotron 2: High-quality direct speech-to-speech translation with voice preservation2021cited by 55 here
  3. Textless Speech-to-Speech Translation on Real Data2021cited by 50 here
  4. CVSS Corpus and Massively Multilingual Speech-to-Speech Translation2022cited by 44 here
  5. UnitY: Two-pass Direct Speech-to-speech Translation with Discrete Units2022cited by 40 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2021Direct speech-to-speech translation with discrete unitscited by 70 here
  2. 2021Translatotron 2: High-quality direct speech-to-speech translation with voice preservationcited by 55 here
  3. 2022CVSS Corpus and Massively Multilingual Speech-to-Speech Translationcited by 44 here
  4. 2022FLEURS: Few-shot Learning Evaluation of Universal Representations of Speechcited by 40 here
  5. 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 100 here
  6. 2023AudioPaLM: A Large Language Model That Can Speak and Listencited by 25 here
18 more it builds on, not in this chain
  1. 2021CTC-based Compression for Direct Speech Translationcited by 23 here
  2. 2021The Multilingual TEDx Corpus for Speech Recognition and Translationcited by 24 here
  3. 2021End-to-end Speech Translation via Cross-modal Progressive Trainingcited by 32 here
  4. 2021Learning Shared Semantic Space for Speech-to-Text Translationcited by 31 here
  5. 2021Stacked Acoustic-and-Textual Encoding: Integrating the Pre-trained Models into Speech Translation Encoderscited by 32 here
  6. 2021Cascade versus Direct Speech Translation: Do the Differences Still Make a Difference?cited by 35 here
  7. 2021RealTranS: End-to-End Simultaneous Speech Translation with Convolutional Weighted-Shrinking Transformercited by 24 here
  8. 2021Direct Simultaneous Speech-to-Text Translation Assisted by Synchronized Streaming ASRcited by 23 here
  9. 2021Improving Speech Translation by Understanding and Learning from the Auxiliary Text Translation Taskcited by 37 here
  10. 2021WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processingcited by 31 here
  11. 2021XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scalecited by 44 here
  12. 2021Textless Speech-to-Speech Translation on Real Datacited by 50 here
  13. 2022STEMM: Self-learning with Speech-text Manifold Mixup for Speech Translationcited by 32 here
  14. 2022Cross-modal Contrastive Learning for Speech Translationcited by 30 here
  15. 2022Over-Generation Cannot Be Rewarded: Length-Adaptive Average Lagging for Simultaneous Speech Translationcited by 31 here
  16. 2022UnitY: Two-pass Direct Speech-to-speech Translation with Discrete Unitscited by 40 here
  17. 2023SeamlessM4T: Massively Multilingual & Multimodal Machine Translationcited by 30 here
  18. 2023Seamless: Multilingual Expressive and Streaming Speech Translationcited by 28 here

The 24 most cited of 739.

Browse

30 of 365, ranked 6–35

  1. 6Improving Speech Translation by Understanding and Learning from the Auxiliary Text Translation Task202137
  2. 7Cascade versus Direct Speech Translation: Do the Differences Still Make a Difference?202135
  3. 8STEMM: Self-learning with Speech-text Manifold Mixup for Speech Translation202232
  4. 9End-to-end Speech Translation via Cross-modal Progressive Training202132
  5. 10Stacked Acoustic-and-Textual Encoding: Integrating the Pre-trained Models into Speech Translation Encoders202132
  6. 11Over-Generation Cannot Be Rewarded: Length-Adaptive Average Lagging for Simultaneous Speech Translation202231
  7. 12Learning Shared Semantic Space for Speech-to-Text Translation202131
  8. 13SeamlessM4T: Massively Multilingual & Multimodal Machine Translation202330
  9. 14Cross-modal Contrastive Learning for Speech Translation202230
  10. 15Seamless: Multilingual Expressive and Streaming Speech Translation202328
All 30
  1. 16RealTranS: End-to-End Simultaneous Speech Translation with Convolutional Weighted-Shrinking Transformer202124
  2. 17Unified Speech-Text Pre-training for Speech Translation and Recognition202223
  3. 18Direct Simultaneous Speech-to-Text Translation Assisted by Synchronized Streaming ASR202123
  4. 19AlignAtt: Using Attention-based Audio-Translation Alignments as a Guide for Simultaneous Speech Translation202322
  5. 20Enhanced Direct Speech-to-Speech Translation Using Self-supervised Pre-training and Data Augmentation202222
  6. 21Fused Acoustic and Text Encoding for Multimodal Bilingual Pretraining and Speech Translation202120
  7. 22CUNI-KIT System for Simultaneous Speech Translation Task at IWSLT 2022202219
  8. 23Revisiting End-to-End Speech-to-Text Translation From Scratch202219
  9. 24SpeechMatrix: A Large-Scale Mined Corpus of Multilingual Speech-to-Speech Translations202219
  10. 25Attention as a Guide for Simultaneous Speech Translation202219
  11. 26mSLAM: Massively multilingual joint pre-training for speech and text202218
  12. 27TranSpeech: Speech-to-Speech Translation With Bilateral Perturbation202218
  13. 28Large-Scale Streaming End-to-End Speech Translation with Neural Transducers202217
  14. 29SHAS: Approaching optimal Segmentation for End-to-End Speech Translation202216
  15. 30Speech Translation with Large Language Models: An Industrial Practice202315
  16. 31StreamSpeech: Simultaneous Speech-to-Speech Translation with Multi-task Learning202414
  17. 32Leveraging unsupervised and weakly-supervised data to improve direct speech-to-speech translation202214
  18. 33Source and Target Bidirectional Knowledge Distillation for End-to-end Speech Translation202114
  19. 34End-to-End Simultaneous Speech Translation with Differentiable Segmentation202313
  20. 35Sample, Translate, Recombine: Leveraging Audio Alignments for Data Augmentation in End-to-end Speech Translation202213