---
title: "Fairseq S2T: Fast Speech-to-Text Modeling with — Wang 2020"
description: "Speech recognition (ASR), speech translation (ST), machine translation (MT) and language modeling (LM) increasingly share data, architectures and…"
authors: ["Changhan Wang", "Yun Tang", "Xutai Ma", "Anne Wu", "Dmytro Okhonko", "Juan Pino"]
year: 2020
venue: "Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing: System Demonstrations"
source: "https://aclanthology.org/2020.aacl-demo.6/"
doi: "10.18653/v1/2020.aacl-demo.6"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/wang-2020-fairseq-s2t-fast-speech-text/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2020-fairseq-s2t-fast-speech-text/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2020](https://md.research.elamur.ai/free/en/papers/2020/page/3/) / Fairseq S2T: Fast Speech-to-Text Modeling with Fairseq

# Fairseq S2T: Fast Speech-to-Text Modeling with Fairseq

**Authors:** [Changhan Wang](https://md.research.elamur.ai/free/en/authors/changhan-wang/), Yun Tang, [Xutai Ma](https://md.research.elamur.ai/free/en/authors/xutai-ma/), [Anne Wu](https://md.research.elamur.ai/free/en/authors/anne-wu/), Dmytro Okhonko, [Juan Pino](https://md.research.elamur.ai/free/en/authors/juan-pino/) · **Year:** 2020 · **Venue:** Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing: System Demonstrations · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2020.aacl-demo.6/) · [PDF](https://aclanthology.org/2020.aacl-demo.6.pdf) · [DOI 10.18653/v1/2020.aacl-demo.6](https://doi.org/10.18653/v1/2020.aacl-demo.6) · [OpenAlex](https://openalex.org/W3092424727)

**Citations:** 122 citations (openalex-snapshot · 2026-10-14) · [137](#cited-by) in our library, of the 140884 papers whose bibliographies we read

## Summary
Speech recognition (ASR), speech translation (ST), machine translation (MT) and language modeling (LM) increasingly share data, architectures and training signals. Researchers therefore need one scalable toolkit that covers all of them. The authors present FAIRSEQ S2T, an extension of the PyTorch-based FAIRSEQ library for end-to-end speech-to-text work. It provides a complete pipeline:
- feature extraction and on-the-fly data transforms;
- training of RNN, Transformer and CTC models;
- offline and simultaneous (streaming) decoding;
- evaluation with built-in scorers and VizSeq visualization.

Because it lives inside FAIRSEQ, existing MT models and LMs can be reused for transfer or multi-task learning. The authors benchmark the toolkit on LibriSpeech (ASR), MuST-C and CoVoST 2 (multilingual ST) using one set of default hyperparameters. The results are close to state-of-the-art systems without per-task tuning. Multilingual ST models clearly beat bilingual ones, and wav2vec features help low-resource directions. Training recipes are released openly, which makes the work a reproducible baseline platform for speech research.

## Key points
- **Data and configuration:** Data is described through TSV manifest files, which work much like Kaldi scp files. Transforms, tokenizer and vocabulary are set in a separate YAML config.
- **Pre-processing and text:** Kaldi-compatible features are computed via PyKaldi or torchaudio, with optional ZIP packing for faster I/O. Online CMVN, speed perturbation and SpecAugment are supported. Tokenizers include Moses, SentencePiece, subword-nmt, byte-level BPE and bytes.
- **Simultaneous ST:** Online models support monotonic attention, wait-k, MILk and monotonic multihead attention. Latency is measured through SimulEval with the AL and DAL metrics.
- **LibriSpeech results:** The best model is T-Lg (263M parameters), with a test WER of 3.5 on clean and 7.8 on other. The RNN B-Big model (52M) reaches 3.9 and 11.5.
- **ST benchmarks:**
  - MuST-C covers about 500 h of TED speech translated into 8 languages.
  - CoVoST 2 contains 2,880 h of speech in 22 languages, with 21 X→En and 15 En→X directions; 13 languages were evaluated.
  - Transformers beat RNNs on every CoVoST 2 direction.
  - Multilingual training with temperature resampling (T=2) gains at least 4 BLEU on Es-En.

## Interesting ideas
- **One default configuration for every benchmark.** Using the same hyperparameters and learning-rate schedule everywhere turns results into a fair toolkit baseline rather than a tuning contest. This would be a useful convention for Ukrainian NLP baselines.
- **Separating data listing from processing.** Plain TSV manifests say what the data is, while a YAML file says how to process it. This makes experiments easy to reproduce and share, and the pattern suits any corpus pipeline.
- **Pre-training the ST encoder on ASR.** Training the translation model's encoder first on high-resource English speech recognition is a cheap way to fight overfitting when labelled data is scarce. The same trick could apply to Ukrainian speech tasks.
- **Self-supervised features transfer across languages.** wav2vec features learned only from English still improved low-resource non-English directions. This suggests that English-pretrained speech representations are a reasonable starting point for under-resourced languages.
- **Reporting simultaneous models by latency band.** Grouping streaming results into high, medium and low latency regimes (by AL) makes the quality-versus-delay trade-off easy to read. This evaluation design fits any real-time system.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Natural language processing](https://md.research.elamur.ai/free/en/shelves/nlp/)

Topics: [Machine translation](https://md.research.elamur.ai/free/en/topics/machine-translation/)

## Cited by

Cited by

137 papers of our library cite it (of the 140884 whose bibliographies we read); 6 of them have a page here, 131 not yet.

- [FINDINGS OF THE IWSLT 2023 EVALUATION CAMPAIGN](https://md.research.elamur.ai/free/en/papers/agarwal-2023-findings-iwslt-2023-evaluation-campaign/)
- [Findings of the IWSLT 2022 Evaluation Campaign](https://md.research.elamur.ai/free/en/papers/anastasopoulos-2022-findings-iwslt-2022-evaluation-campaign/)
- [English to Central Kurdish Speech Translation: Corpus Creation, Evaluation, and Orthographic Standardization](https://md.research.elamur.ai/free/en/papers/mohammadamini-2026-english-central-kurdish-speech-translation/)
- [LIUM Submission for IWSLT 2026 Low-resource Speech Translation Track](https://md.research.elamur.ai/free/en/papers/mohammadamini-2026-lium-submission-iwslt-2026-low/)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://md.research.elamur.ai/free/en/papers/radford-2022-robust-speech-recognition-large-scale/)
- [VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation](https://md.research.elamur.ai/free/en/papers/wang-2021-voxpopuli-large-scale-multilingual-speech/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{wang2020fairseq,
  title = {{Fairseq S2T: Fast Speech-to-Text Modeling with Fairseq}},
  author = {Wang, Changhan and Tang, Yun and Ma, Xutai and Wu, Anne and Okhonko, Dmytro and Pino, Juan},
  booktitle = {Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing: System Demonstrations},
  year = {2020},
  doi = {10.18653/v1/2020.aacl-demo.6},
  url = {https://doi.org/10.18653/v1/2020.aacl-demo.6}
}
```

- [← previous Factual Error Correction for Abstractive Summarization Models](https://md.research.elamur.ai/free/en/papers/cao-2020-factual-error-correction-abstractive-summarization/)
- [next → Fast and Accurate Non-Projective Dependency Tree Linearization](https://md.research.elamur.ai/free/en/papers/yu-2020-fast-accurate-non-projective-dependency/)
- [↑ back to papers of 2020](https://md.research.elamur.ai/free/en/papers/2020/page/3/)

## References

58 references, 18 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Antonios Anastasopoulos and David Chiang. 2018. Tied multitask learning for neural speech translation. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, From a wav2vec model pre-trained on LibriSpeech: h… [find](https://scholar.google.com/scholar?q=Antonios+Anastasopoulos+and+David+Chiang%2E+2018%2E+Tied+multitask+learning+for+neural+speech+translation%2E+In+Proceedings+of+the+2018+Conference+of+the+North+American+Chapter+of+the+Association+for+Comput) [our card](https://md.research.elamur.ai/free/en/papers/anastasopoulos-2018-tied-multitask-learning-neural-speech/)
2. Ebrahim Ansari, Amittai Axelrod, Nguyen Bach, Ondřej Bojar, Roldano Cattoni, Fahim Dalvi, Nadir Durrani, Marcello Federico, Christian Federmann, Jiatao Gu, Fei Huang, Kevin Knight, Xutai Ma, Ajay Nagesh, Matteo Negri, Jan Niehues, Juan Pino, Elizabeth Salesky, Xing Shi, Sebastian Stüker, Marco Tur… [find](https://scholar.google.com/scholar?q=Ebrahim+Ansari%2C+Amittai+Axelrod%2C+Nguyen+Bach%2C+Ondr%CC%8Cej+Bojar%2C+Roldano+Cattoni%2C+Fahim+Dalvi%2C+Nadir+Durrani%2C+Marcello+Federico%2C+Christian+Federmann%2C+Jiatao+Gu%2C+Fei+Huang%2C+Kevin+Knight%2C+Xutai+Ma%2C+Ajay+Nag)
3. FINDINGS OF THE IWSLT 2020 EVALU-ATION CAMPAIGN. In Proceedings of the 17th International Conference on Spoken Language Translation, pages 1–34, Online. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=FINDINGS+OF+THE+IWSLT+2020+EVALU%2DATION+CAMPAIGN%2E+In+Proceedings+of+the+17th+International+Conference+on+Spoken+Language+Translation%2C+pages+1%E2%80%9334%2C+Online%2E+Association+for+Computational+Linguistics%2E)
4. Colin Cherry and George Foster. 2019. Thinking slow about latency evaluation for simultaneous machine translation. arXiv preprint arXiv:1906.00048. [find](https://scholar.google.com/scholar?q=Colin+Cherry+and+George+Foster%2E+2019%2E+Thinking+slow+about+latency+evaluation+for+simultaneous+machine+translation%2E+arXiv+preprint+arXiv%3A1906%2E00048%2E)
5. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. In Proceedings of the 58th Annual Meeting of the Association f… [find](https://scholar.google.com/scholar?q=Alexis+Conneau%2C+Kartikay+Khandelwal%2C+Naman+Goyal%2C+Vishrav+Chaudhary%2C+Guillaume+Wenzek%2C+Francisco+Guzma%CC%81n%2C+Edouard+Grave%2C+Myle+Ott%2C+Luke+Zettlemoyer%2C+and+Veselin+Stoyanov%2E+2020%2E+Unsupervised+cross%2Dling)
6. Naveen Arivazhagan, Ankur Bapna, Orhan Firat, Dmitry Lepikhin, Melvin Johnson, Maxim Krikun, Mia Xu Chen, Yuan Cao, George Foster, Colin Cherry, et al. 2019a. Massively multilingual neural machine translation in the wild: Findings and challenges. arXiv preprint arXiv:1907.05019. [find](https://scholar.google.com/scholar?q=Naveen+Arivazhagan%2C+Ankur+Bapna%2C+Orhan+Firat%2C+Dmitry+Lepikhin%2C+Melvin+Johnson%2C+Maxim+Krikun%2C+Mia+Xu+Chen%2C+Yuan+Cao%2C+George+Foster%2C+Colin+Cherry%2C+et+al%2E+2019a%2E+Massively+multilingual+neural+machine+tra) [our card](https://md.research.elamur.ai/free/en/papers/arivazhagan-2019-massively-multilingual-neural-machine-translation/)
7. M. A. Di Gangi, M. Negri, and M. Turchi. 2019. Oneto-many multilingual end-to-end speech translation. In 2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), pages 585–592. [find](https://scholar.google.com/scholar?q=M%2E+A%2E+Di+Gangi%2C+M%2E+Negri%2C+and+M%2E+Turchi%2E+2019%2E+Oneto%2Dmany+multilingual+end%2Dto%2Dend+speech+translation%2E+In+2019+IEEE+Automatic+Speech+Recognition+and+Understanding+Workshop+%28ASRU%29%2C+pages+585%E2%80%93592%2E)
8. Mattia A Di Gangi, Roldano Cattoni, Luisa Bentivogli, Matteo Negri, and Marco Turchi. 2019a. Must-c: a multilingual speech translation corpus. In 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2012–2017. Association… [find](https://scholar.google.com/scholar?q=Mattia+A+Di+Gangi%2C+Roldano+Cattoni%2C+Luisa+Bentivogli%2C+Matteo+Negri%2C+and+Marco+Turchi%2E+2019a%2E+Must%2Dc%3A+a+multilingual+speech+translation+corpus%2E+In+2019+Conference+of+the+North+American+Chapter+of+the+A) [our card](https://md.research.elamur.ai/free/en/papers/gangi-2019-must-c-multilingual-speech-translation/)
9. Naveen Arivazhagan, Colin Cherry, Wolfgang Macherey, Chung-Cheng Chiu, Semih Yavuz, Ruoming Pang, Wei Li, and Colin Raffel. 2019b. Monotonic infinite lookback attention for simultaneous machine translation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, p… [find](https://scholar.google.com/scholar?q=Naveen+Arivazhagan%2C+Colin+Cherry%2C+Wolfgang+Macherey%2C+Chung%2DCheng+Chiu%2C+Semih+Yavuz%2C+Ruoming+Pang%2C+Wei+Li%2C+and+Colin+Raffel%2E+2019b%2E+Monotonic+infinite+lookback+attention+for+simultaneous+machine+transl) [our card](https://md.research.elamur.ai/free/en/papers/arivazhagan-2019-monotonic-infinite-lookback-attention-simultaneous/)
10. Mattia Antonino Di Gangi, Matteo Negri, Roldano Cattoni, Roberto Dessi, and Marco Turchi. 2019b. Enhancing transformer for end-to-end speech-to-text translation. In Proceedings of Machine Translation Summit XVII Volume 1: Research Track, pages 21– 31, Dublin, Ireland. European Association for Machin… [find](https://scholar.google.com/scholar?q=Mattia+Antonino+Di+Gangi%2C+Matteo+Negri%2C+Roldano+Cattoni%2C+Roberto+Dessi%2C+and+Marco+Turchi%2E+2019b%2E+Enhancing+transformer+for+end%2Dto%2Dend+speech%2Dto%2Dtext+translation%2E+In+Proceedings+of+Machine+Translation+)
11. Sameer Bansal, Herman Kamper, Karen Livescu, Adam Lopez, and Sharon Goldwater. 2019. Pre-training on high-resource speech recognition improves low-resource speech-to-text translation. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics… [find](https://scholar.google.com/scholar?q=Sameer+Bansal%2C+Herman+Kamper%2C+Karen+Livescu%2C+Adam+Lopez%2C+and+Sharon+Goldwater%2E+2019%2E+Pre%2Dtraining+on+high%2Dresource+speech+recognition+improves+low%2Dresource+speech%2Dto%2Dtext+translation%2E+In+Proceedings+o)
12. Long Duong, Antonios Anastasopoulos, David Chiang, Steven Bird, and Trevor Cohn. 2016. An attentional model for speech translation without transcription. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,… [find](https://scholar.google.com/scholar?q=Long+Duong%2C+Antonios+Anastasopoulos%2C+David+Chiang%2C+Steven+Bird%2C+and+Trevor+Cohn%2E+2016%2E+An+attentional+model+for+speech+translation+without+transcription%2E+In+Proceedings+of+the+2016+Conference+of+the+N)
13. Alexandre Bérard, Laurent Besacier, Ali Can Kocabiyikoglu, and Olivier Pietquin. 2018. End-to-end automatic speech translation of audiobooks. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6224–6228. IEEE. [find](https://scholar.google.com/scholar?q=Alexandre+Be%CC%81rard%2C+Laurent+Besacier%2C+Ali+Can+Kocabiyikoglu%2C+and+Olivier+Pietquin%2E+2018%2E+End%2Dto%2Dend+automatic+speech+translation+of+audiobooks%2E+In+2018+IEEE+International+Conference+on+Acoustics%2C+Speec)
14. Alex Graves, Santiago Fernández, Faustino Gomez, and Jürgen Schmidhuber. 2006. Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks. In Proceedings of the 23rd international conference on Machine learning, pages 369–376. [find](https://scholar.google.com/scholar?q=Alex+Graves%2C+Santiago+Ferna%CC%81ndez%2C+Faustino+Gomez%2C+and+Ju%CC%88rgen+Schmidhuber%2E+2006%2E+Connectionist+temporal+classification%3A+labelling+unsegmented+sequence+data+with+recurrent+neural+networks%2E+In+Proceedin)
15. Alexandre Bérard, Olivier Pietquin, Christophe Servan, and Laurent Besacier. 2016. Listen and translate: A proof of concept for end-to-end speech-to-text translation. arXiv preprint arXiv:1612.01744. — [arXiv:1612.01744](https://arxiv.org/abs/1612.01744) [find](https://scholar.google.com/scholar?q=Alexandre+Be%CC%81rard%2C+Olivier+Pietquin%2C+Christophe+Servan%2C+and+Laurent+Besacier%2E+2016%2E+Listen+and+translate%3A+A+proof+of+concept+for+end%2Dto%2Dend+speech%2Dto%2Dtext+translation%2E+arXiv+preprint+arXiv%3A1612%2E01744%2E)
16. Charles R Harris, K Jarrod Millman, Stéfan J van der Walt, Ralf Gommers, Pauli Virtanen, David Cournapeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J Smith, et al. 2020. Array programming with numpy. Nature, 585(7825):357–362. [find](https://scholar.google.com/scholar?q=Charles+R+Harris%2C+K+Jarrod+Millman%2C+Ste%CC%81fan+J+van+der+Walt%2C+Ralf+Gommers%2C+Pauli+Virtanen%2C+David+Cournapeau%2C+Eric+Wieser%2C+Julian+Taylor%2C+Sebastian+Berg%2C+Nathaniel+J+Smith%2C+et+al%2E+2020%2E+Array+programmin) [our card](https://md.research.elamur.ai/free/en/papers/harris-2020-array-programming-numpy/)
17. Dogan Can, Victor R. Martinez, Pavlos Papadopoulos, and Shrikanth S. Narayanan. 2018. Pykaldi: A python wrapper for kaldi. In Acoustics, Speech and Signal Processing (ICASSP), 2018 IEEE International Conference on. IEEE. [find](https://scholar.google.com/scholar?q=Dogan+Can%2C+Victor+R%2E+Martinez%2C+Pavlos+Papadopoulos%2C+and+Shrikanth+S%2E+Narayanan%2E+2018%2E+Pykaldi%3A+A+python+wrapper+for+kaldi%2E+In+Acoustics%2C+Speech+and+Signal+Processing+%28ICASSP%29%2C+2018+IEEE+International+)
18. Hirofumi Inaguma, Shun Kiyono, Kevin Duh, Shigeki Karita, Nelson Yalta, Tomoki Hayashi, and Shinji Watanabe. 2020. ESPnet-ST: All-in-one speech translation toolkit. In Proceedings of the 58th Annual Meeting of the Association for Computational William Chan, Navdeep Jaitly, Quoc Le, and Oriol Vinyals… [find](https://scholar.google.com/scholar?q=Hirofumi+Inaguma%2C+Shun+Kiyono%2C+Kevin+Duh%2C+Shigeki+Karita%2C+Nelson+Yalta%2C+Tomoki+Hayashi%2C+and+Shinji+Watanabe%2E+2020%2E+ESPnet%2DST%3A+All%2Din%2Done+speech+translation+toolkit%2E+In+Proceedings+of+the+58th+Annual+M)
19. Linguistics: System Demonstrations, pages 302– 311, Online. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Linguistics%3A+System+Demonstrations%2C+pages+302%E2%80%93+311%2C+Online%2E+Association+for+Computational+Linguistics%2E)
20. Jacob Kahn, Ann Lee, and Awni Hannun. 2020. Self-training for end-to-end speech recognition. [find](https://scholar.google.com/scholar?q=Jacob+Kahn%2C+Ann+Lee%2C+and+Awni+Hannun%2E+2020%2E+Self%2Dtraining+for+end%2Dto%2Dend+speech+recognition%2E)
21. ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 7084–7088. [find](https://scholar.google.com/scholar?q=ICASSP+2020+%2D+2020+IEEE+International+Conference+on+Acoustics%2C+Speech+and+Signal+Processing+%28ICASSP%29%2C+pages+7084%E2%80%937088%2E)
22. Mingbo Ma, Liang Huang, Hao Xiong, Renjie Zheng, Kaibo Liu, Baigong Zheng, Chuanqiang Zhang, Zhongjun He, Hairong Liu, Xing Li, Hua Wu, and Haifeng Wang. 2019. STACL: Simultaneous translation with implicit anticipation and controllable latency using prefix-to-prefix framework. In Proceedings of the… [find](https://scholar.google.com/scholar?q=Mingbo+Ma%2C+Liang+Huang%2C+Hao+Xiong%2C+Renjie+Zheng%2C+Kaibo+Liu%2C+Baigong+Zheng%2C+Chuanqiang+Zhang%2C+Zhongjun+He%2C+Hairong+Liu%2C+Xing+Li%2C+Hua+Wu%2C+and+Haifeng+Wang%2E+2019%2E+STACL%3A+Simultaneous+translation+with+imp) [our card](https://md.research.elamur.ai/free/en/papers/ma-2019-stacl-simultaneous-translation-implicit-anticipation/)
23. Guillaume Klein, Yoon Kim, Yuntian Deng, Jean Senellart, and Alexander Rush. 2017. OpenNMT: Open-source toolkit for neural machine translation. In Proceedings of ACL 2017, System Demonstrations, pages 67–72, Vancouver, Canada. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Guillaume+Klein%2C+Yoon+Kim%2C+Yuntian+Deng%2C+Jean+Senellart%2C+and+Alexander+Rush%2E+2017%2E+OpenNMT%3A+Open%2Dsource+toolkit+for+neural+machine+translation%2E+In+Proceedings+of+ACL+2017%2C+System+Demonstrations%2C+pages) [our card](https://md.research.elamur.ai/free/en/papers/klein-2017-opennmt-open-source-toolkit-neural/)
24. Tom Ko, Vijayaditya Peddinti, Daniel Povey, Michael L Seltzer, and Sanjeev Khudanpur. 2017. A study on data augmentation of reverberant speech for robust speech recognition. In 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 5220–5224. IEEE. [find](https://scholar.google.com/scholar?q=Tom+Ko%2C+Vijayaditya+Peddinti%2C+Daniel+Povey%2C+Michael+L+Seltzer%2C+and+Sanjeev+Khudanpur%2E+2017%2E+A+study+on+data+augmentation+of+reverberant+speech+for+robust+speech+recognition%2E+In+2017+IEEE+International)
25. Xutai Ma, Mohammad Javad Dousti, Changhan Wang, Jiatao Gu, and Juan Pino. 2020a. Simuleval: An evaluation toolkit for simultaneous translation. arXiv preprint arXiv:2007.16193. [find](https://scholar.google.com/scholar?q=Xutai+Ma%2C+Mohammad+Javad+Dousti%2C+Changhan+Wang%2C+Jiatao+Gu%2C+and+Juan+Pino%2E+2020a%2E+Simuleval%3A+An+evaluation+toolkit+for+simultaneous+translation%2E+arXiv+preprint+arXiv%3A2007%2E16193%2E) [our card](https://md.research.elamur.ai/free/en/papers/ma-2020-simuleval-evaluation-toolkit-simultaneous-translation/)
26. Xutai Ma, Juan Pino, James Cross, Liezl Puzon, and Jiatao Gu. 2020b. Monotonic multihead attention. 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020, Conference Track Proceedings. [find](https://scholar.google.com/scholar?q=Xutai+Ma%2C+Juan+Pino%2C+James+Cross%2C+Liezl+Puzon%2C+and+Jiatao+Gu%2E+2020b%2E+Monotonic+multihead+attention%2E+8th+International+Conference+on+Learning+Representations%2C+ICLR+2020%2C+Addis+Ababa%2C+Ethiopia%2C+April+26)
27. Oleksii Kuchaiev, Boris Ginsburg, Igor Gitman, Vitaly Lavrukhin, Carl Case, and Paulius Micikevicius. 2018. Openseq2seq: extensible toolkit for distributed and mixed precision training of sequence-to-sequence models. In Proceedings of Workshop for NLP Open Source Software (NLP-OSS), pages 41– 46. [find](https://scholar.google.com/scholar?q=Oleksii+Kuchaiev%2C+Boris+Ginsburg%2C+Igor+Gitman%2C+Vitaly+Lavrukhin%2C+Carl+Case%2C+and+Paulius+Micikevicius%2E+2018%2E+Openseq2seq%3A+extensible+toolkit+for+distributed+and+mixed+precision+training+of+sequence%2Dto%2D)
28. Paulius Micikevicius, Sharan Narang, Jonah Alben, Gregory Diamos, Erich Elsen, David Garcia, Boris Ginsburg, Michael Houston, Oleksii Kuchaiev, Ganesh Venkatesh, et al. 2018. Mixed precision training. In International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Paulius+Micikevicius%2C+Sharan+Narang%2C+Jonah+Alben%2C+Gregory+Diamos%2C+Erich+Elsen%2C+David+Garcia%2C+Boris+Ginsburg%2C+Michael+Houston%2C+Oleksii+Kuchaiev%2C+Ganesh+Venkatesh%2C+et+al%2E+2018%2E+Mixed+precision+training%2E)
29. Taku Kudo and John Richardson. 2018. SentencePiece: A simple and language independent subword tokenizer and detokenizer for neural text processing. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 66–71, Brussels, Belgium. Associ… [find](https://scholar.google.com/scholar?q=Taku+Kudo+and+John+Richardson%2E+2018%2E+SentencePiece%3A+A+simple+and+language+independent+subword+tokenizer+and+detokenizer+for+neural+text+processing%2E+In+Proceedings+of+the+2018+Conference+on+Empirical+M) [our card](https://md.research.elamur.ai/free/en/papers/kudo-2018-sentencepiece-simple-language-independent-subword/)
30. Abdelrahman Mohamed, Dmytro Okhonko, and Luke Zettlemoyer. 2019. Transformers with convolutional context for asr. arXiv preprint arXiv:1904.11660. — [arXiv:1904.11660](https://arxiv.org/abs/1904.11660) [find](https://scholar.google.com/scholar?q=Abdelrahman+Mohamed%2C+Dmytro+Okhonko%2C+and+Luke+Zettlemoyer%2E+2019%2E+Transformers+with+convolutional+context+for+asr%2E+arXiv+preprint+arXiv%3A1904%2E11660%2E+%E2%80%94+%5BarXiv%3A1904%2E11660%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1904%2E11660)
31. Nathan Ng, Kyra Yee, Alexei Baevski, Myle Ott, Michael Auli, and Sergey Edunov. 2019. Facebook FAIR’s WMT19 news translation task submission. In Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1), pages 314–319, Florence, Italy. Association for Computat… [find](https://scholar.google.com/scholar?q=Nathan+Ng%2C+Kyra+Yee%2C+Alexei+Baevski%2C+Myle+Ott%2C+Michael+Auli%2C+and+Sergey+Edunov%2E+2019%2E+Facebook+FAIR%E2%80%99s+WMT19+news+translation+task+submission%2E+In+Proceedings+of+the+Fourth+Conference+on+Machine+Transla) [our card](https://md.research.elamur.ai/free/en/papers/ng-2019-facebook-fair-s-wmt19-news/)
32. Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Veselin Stoyanov, and Luke Zettlemoyer. 2020. BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. In Proceedings of the 58th Annual Meeting of the… [find](https://scholar.google.com/scholar?q=Mike+Lewis%2C+Yinhan+Liu%2C+Naman+Goyal%2C+Marjan+Ghazvininejad%2C+Abdelrahman+Mohamed%2C+Omer+Levy%2C+Veselin+Stoyanov%2C+and+Luke+Zettlemoyer%2E+2020%2E+BART%3A+Denoising+sequence%2Dto%2Dsequence+pre%2Dtraining+for+natural+l) [our card](https://md.research.elamur.ai/free/en/papers/lewis-2020-bart-denoising-sequence-sequence-pre/)
33. Myle Ott, Sergey Edunov, Alexei Baevski, Angela Fan, Sam Gross, Nathan Ng, David Grangier, and Michael Auli. 2019. fairseq: A fast, extensible toolkit for sequence modeling. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics (Demonstr… [find](https://scholar.google.com/scholar?q=Myle+Ott%2C+Sergey+Edunov%2C+Alexei+Baevski%2C+Angela+Fan%2C+Sam+Gross%2C+Nathan+Ng%2C+David+Grangier%2C+and+Michael+Auli%2E+2019%2E+fairseq%3A+A+fast%2C+extensible+toolkit+for+sequence+modeling%2E+In+Proceedings+of+the+2019) [our card](https://md.research.elamur.ai/free/en/papers/ott-2019-fairseq-fast-extensible-toolkit-sequence/)
34. Bo Li, Yu Zhang, Tara Sainath, Yonghui Wu, and William Chan. 2019. Bytes are all you need: End-to-end multilingual speech recognition and synthesis with bytes. In ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 5621–5625. IEEE. [find](https://scholar.google.com/scholar?q=Bo+Li%2C+Yu+Zhang%2C+Tara+Sainath%2C+Yonghui+Wu%2C+and+William+Chan%2E+2019%2E+Bytes+are+all+you+need%3A+End%2Dto%2Dend+multilingual+speech+recognition+and+synthesis+with+bytes%2E+In+ICASSP+2019%2D2019+IEEE+International+C)
35. Vassil Panayotov, Guoguo Chen, Daniel Povey, and Sanjeev Khudanpur. 2015. Librispeech: an asr corpus based on public domain audio books. In 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 5206–5210. IEEE. [find](https://scholar.google.com/scholar?q=Vassil+Panayotov%2C+Guoguo+Chen%2C+Daniel+Povey%2C+and+Sanjeev+Khudanpur%2E+2015%2E+Librispeech%3A+an+asr+corpus+based+on+public+domain+audio+books%2E+In+2015+IEEE+International+Conference+on+Acoustics%2C+Speech+and+)
36. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692. [find](https://scholar.google.com/scholar?q=Yinhan+Liu%2C+Myle+Ott%2C+Naman+Goyal%2C+Jingfei+Du%2C+Mandar+Joshi%2C+Danqi+Chen%2C+Omer+Levy%2C+Mike+Lewis%2C+Luke+Zettlemoyer%2C+and+Veselin+Stoyanov%2E+2019%2E+Roberta%3A+A+robustly+optimized+bert+pretraining+approach%2E+a)
37. Gabriel Synnaeve, Qiantong Xu, Jacob Kahn, Edouard Grave, Tatiana Likhomanenko, Vineel Pratap, Anuroop Sriram, Vitaliy Liptchinsky, and Ronan Collobert. 2019. End-to-end asr: from supervised to semi-supervised learning with modern architectures. arXiv preprint arXiv:1911.08460. — [arXiv:1911.08460](https://arxiv.org/abs/1911.08460) [find](https://scholar.google.com/scholar?q=Gabriel+Synnaeve%2C+Qiantong+Xu%2C+Jacob+Kahn%2C+Edouard+Grave%2C+Tatiana+Likhomanenko%2C+Vineel+Pratap%2C+Anuroop+Sriram%2C+Vitaliy+Liptchinsky%2C+and+Ronan+Collobert%2E+2019%2E+End%2Dto%2Dend+asr%3A+from+supervised+to+semi%2Ds)
38. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting of the Association for Computational Linguistics, pages 311–318. [find](https://scholar.google.com/scholar?q=Kishore+Papineni%2C+Salim+Roukos%2C+Todd+Ward%2C+and+Wei%2DJing+Zhu%2E+2002%2E+Bleu%3A+a+method+for+automatic+evaluation+of+machine+translation%2E+In+Proceedings+of+the+40th+annual+meeting+of+the+Association+for+Comp) [our card](https://md.research.elamur.ai/free/en/papers/papineni-2002-bleu-method-automatic-evaluation-machine/)
39. Daniel S. Park, William Chan, Yu Zhang, Chung-Cheng Chiu, Barret Zoph, Ekin D. Cubuk, and Quoc V. Le. 2019. Specaugment: A simple data augmentation method for automatic speech recognition. [find](https://scholar.google.com/scholar?q=Daniel+S%2E+Park%2C+William+Chan%2C+Yu+Zhang%2C+Chung%2DCheng+Chiu%2C+Barret+Zoph%2C+Ekin+D%2E+Cubuk%2C+and+Quoc+V%2E+Le%2E+2019%2E+Specaugment%3A+A+simple+data+augmentation+method+for+automatic+speech+recognition%2E) [our card](https://md.research.elamur.ai/free/en/papers/park-2019-specaugment-simple-data-augmentation-method/)
40. Ashish Vaswani, Samy Bengio, Eugene Brevdo, Francois Chollet, Aidan Gomez, Stephan Gouws, Llion Jones, Łukasz Kaiser, Nal Kalchbrenner, Niki Parmar, Ryan Sepassi, Noam Shazeer, and Jakob Uszkoreit. 2018. Tensor2Tensor for neural machine translation. In Proceedings of the 13th Conference of the Assoc… [find](https://scholar.google.com/scholar?q=Ashish+Vaswani%2C+Samy+Bengio%2C+Eugene+Brevdo%2C+Francois+Chollet%2C+Aidan+Gomez%2C+Stephan+Gouws%2C+Llion+Jones%2C+%C5%81ukasz+Kaiser%2C+Nal+Kalchbrenner%2C+Niki+Parmar%2C+Ryan+Sepassi%2C+Noam+Shazeer%2C+and+Jakob+Uszkoreit%2E+20) [our card](https://md.research.elamur.ai/free/en/papers/vaswani-2018-tensor2tensor-neural-machine-translation/)
41. Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. 2019. Pytorch: An imperative style, high-performance deep learning library. In Advances in neural information processing systems, pages 8026–8037. [find](https://scholar.google.com/scholar?q=Adam+Paszke%2C+Sam+Gross%2C+Francisco+Massa%2C+Adam+Lerer%2C+James+Bradbury%2C+Gregory+Chanan%2C+Trevor+Killeen%2C+Zeming+Lin%2C+Natalia+Gimelshein%2C+Luca+Antiga%2C+et+al%2E+2019%2E+Pytorch%3A+An+imperative+style%2C+high%2Dperfor) [our card](https://md.research.elamur.ai/free/en/papers/paszke-2019-pytorch-imperative-style-high-performance/)
42. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In Advances in neural information processing systems, pages 5998–6008. [find](https://scholar.google.com/scholar?q=Ashish+Vaswani%2C+Noam+Shazeer%2C+Niki+Parmar%2C+Jakob+Uszkoreit%2C+Llion+Jones%2C+Aidan+N+Gomez%2C+%C5%81ukasz+Kaiser%2C+and+Illia+Polosukhin%2E+2017%2E+Attention+is+all+you+need%2E+In+Advances+in+neural+information+processi)
43. Juan Pino, Qiantong Xu, Xutai Ma, Mohammad Javad Dousti, and Yun Tang. 2020. Self-training for end-to-end speech translation. arXiv preprint arXiv:2006.02490. — [arXiv:2006.02490](https://arxiv.org/abs/2006.02490) [find](https://scholar.google.com/scholar?q=Juan+Pino%2C+Qiantong+Xu%2C+Xutai+Ma%2C+Mohammad+Javad+Dousti%2C+and+Yun+Tang%2E+2020%2E+Self%2Dtraining+for+end%2Dto%2Dend+speech+translation%2E+arXiv+preprint+arXiv%3A2006%2E02490%2E+%E2%80%94+%5BarXiv%3A2006%2E02490%5D%28https%3A%2F%2Farxiv%2Eorg%2Fab)
44. Maja Popović. 2015. chrf: character n-gram f-score for automatic mt evaluation. In Proceedings of the Tenth Workshop on Statistical Machine Translation, pages 392–395. [find](https://scholar.google.com/scholar?q=Maja+Popovic%CC%81%2E+2015%2E+chrf%3A+character+n%2Dgram+f%2Dscore+for+automatic+mt+evaluation%2E+In+Proceedings+of+the+Tenth+Workshop+on+Statistical+Machine+Translation%2C+pages+392%E2%80%93395%2E) [our card](https://md.research.elamur.ai/free/en/papers/popovic-2015-chrf-character-n-gram-f/)
45. Changhan Wang, Kyunghyun Cho, and Jiatao Gu. 2020a. Neural machine translation with byte-level subwords. Proceedings of the AAAI Conference on Artificial Intelligence, 34:9154–9160. [find](https://scholar.google.com/scholar?q=Changhan+Wang%2C+Kyunghyun+Cho%2C+and+Jiatao+Gu%2E+2020a%2E+Neural+machine+translation+with+byte%2Dlevel+subwords%2E+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelligence%2C+34%3A9154%E2%80%939160%2E)
46. Matt Post. 2018. A call for clarity in reporting BLEU scores. In Proceedings of the Third Conference on Machine Translation: Research Papers, pages 186– 191, Belgium, Brussels. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Matt+Post%2E+2018%2E+A+call+for+clarity+in+reporting+BLEU+scores%2E+In+Proceedings+of+the+Third+Conference+on+Machine+Translation%3A+Research+Papers%2C+pages+186%E2%80%93+191%2C+Belgium%2C+Brussels%2E+Association+for+Computa) [our card](https://md.research.elamur.ai/free/en/papers/post-2018-call-clarity-reporting-bleu-scores/)
47. Changhan Wang, Anirudh Jain, Danlu Chen, and Jiatao Gu. 2019. VizSeq: a visual analysis toolkit for text generation tasks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP… [find](https://scholar.google.com/scholar?q=Changhan+Wang%2C+Anirudh+Jain%2C+Danlu+Chen%2C+and+Jiatao+Gu%2E+2019%2E+VizSeq%3A+a+visual+analysis+toolkit+for+text+generation+tasks%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Methods+in+Natural+Languag)
48. Daniel Povey, Arnab Ghoshal, Gilles Boulianne, Lukas Burget, Ondrej Glembek, Nagendra Goel, Mirko Hannemann, Petr Motlicek, Yanmin Qian, Petr Schwarz, et al. 2011. The kaldi speech recognition toolkit. In IEEE 2011 workshop on automatic speech recognition and understanding, CONF. IEEE Signal Process… [find](https://scholar.google.com/scholar?q=Daniel+Povey%2C+Arnab+Ghoshal%2C+Gilles+Boulianne%2C+Lukas+Burget%2C+Ondrej+Glembek%2C+Nagendra+Goel%2C+Mirko+Hannemann%2C+Petr+Motlicek%2C+Yanmin+Qian%2C+Petr+Schwarz%2C+et+al%2E+2011%2E+The+kaldi+speech+recognition+toolkit)
49. Changhan Wang, Juan Pino, and Jiatao Gu. 2020b. Improving cross-lingual transfer learning for end-to-end speech recognition with speech translation. arXiv preprint arXiv:2006.05474. — [arXiv:2006.05474](https://arxiv.org/abs/2006.05474) [find](https://scholar.google.com/scholar?q=Changhan+Wang%2C+Juan+Pino%2C+and+Jiatao+Gu%2E+2020b%2E+Improving+cross%2Dlingual+transfer+learning+for+end%2Dto%2Dend+speech+recognition+with+speech+translation%2E+arXiv+preprint+arXiv%3A2006%2E05474%2E+%E2%80%94+%5BarXiv%3A2006%2E0547)
50. Vineel Pratap, Awni Hannun, Qiantong Xu, Jeff Cai, Jacob Kahn, Gabriel Synnaeve, Vitaliy Liptchinsky, and Ronan Collobert. 2018. wav2letter++: The fastest open-source speech recognition system. CoRR, abs/1812.07625. [find](https://scholar.google.com/scholar?q=Vineel+Pratap%2C+Awni+Hannun%2C+Qiantong+Xu%2C+Jeff+Cai%2C+Jacob+Kahn%2C+Gabriel+Synnaeve%2C+Vitaliy+Liptchinsky%2C+and+Ronan+Collobert%2E+2018%2E+wav2letter%2B%2B%3A+The+fastest+open%2Dsource+speech+recognition+system%2E+CoRR%2C+)
51. Changhan Wang, Anne Wu, and Juan Pino. 2020c. Covost 2 and massively multilingual speech-to-text translation. arXiv e-prints, pages arXiv–2007. [find](https://scholar.google.com/scholar?q=Changhan+Wang%2C+Anne+Wu%2C+and+Juan+Pino%2E+2020c%2E+Covost+2+and+massively+multilingual+speech%2Dto%2Dtext+translation%2E+arXiv+e%2Dprints%2C+pages+arXiv%E2%80%932007%2E)
52. Anne Wu, Changhan Wang, Juan Pino, and Jiatao Gu. 2020. Self-supervised representations improve end-to-end speech translation. arXiv preprint arXiv:2006.12124. — [arXiv:2006.12124](https://arxiv.org/abs/2006.12124) [find](https://scholar.google.com/scholar?q=Anne+Wu%2C+Changhan+Wang%2C+Juan+Pino%2C+and+Jiatao+Gu%2E+2020%2E+Self%2Dsupervised+representations+improve+end%2Dto%2Dend+speech+translation%2E+arXiv+preprint+arXiv%3A2006%2E12124%2E+%E2%80%94+%5BarXiv%3A2006%2E12124%5D%28https%3A%2F%2Farxiv%2Eorg%2Fa)
53. Colin Raffel, Minh-Thang Luong, Peter J Liu, Ron J Weiss, and Douglas Eck. 2017. Online and linear-time attention by enforcing monotonic alignments. In International Conference on Machine Learning, pages 2837–2846. [find](https://scholar.google.com/scholar?q=Colin+Raffel%2C+Minh%2DThang+Luong%2C+Peter+J+Liu%2C+Ron+J+Weiss%2C+and+Douglas+Eck%2E+2017%2E+Online+and+linear%2Dtime+attention+by+enforcing+monotonic+alignments%2E+In+International+Conference+on+Machine+Learning%2C+pa)
54. Thomas Zenkel, Matthias Sperber, Jan Niehues, Markus Müller, Ngoc-Quan Pham, Sebastian Stüker, and Alex Waibel. 2018. Open source toolkit for speech to text translation. The Prague Bulletin of Mathematical Linguistics, 111(1):125–135. [find](https://scholar.google.com/scholar?q=Thomas+Zenkel%2C+Matthias+Sperber%2C+Jan+Niehues%2C+Markus+Mu%CC%88ller%2C+Ngoc%2DQuan+Pham%2C+Sebastian+Stu%CC%88ker%2C+and+Alex+Waibel%2E+2018%2E+Open+source+toolkit+for+speech+to+text+translation%2E+The+Prague+Bulletin+of+Mathe)
55. Steffen Schneider, Alexei Baevski, Ronan Collobert, and Michael Auli. 2019. wav2vec: Unsupervised Pre-Training for Speech Recognition. In Proc. [find](https://scholar.google.com/scholar?q=Steffen+Schneider%2C+Alexei+Baevski%2C+Ronan+Collobert%2C+and+Michael+Auli%2E+2019%2E+wav2vec%3A+Unsupervised+Pre%2DTraining+for+Speech+Recognition%2E+In+Proc%2E)
56. Interspeech 2019, pages 3465–3469. [find](https://scholar.google.com/scholar?q=Interspeech+2019%2C+pages+3465%E2%80%933469%2E)
57. Albert Zeyer, Tamer Alkhouli, and Hermann Ney. 2018. RETURNN as a generic flexible neural toolkit with application to translation and speech recognition. In Proceedings of ACL 2018, System Demonstrations, pages 128–133, Melbourne, Australia. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Albert+Zeyer%2C+Tamer+Alkhouli%2C+and+Hermann+Ney%2E+2018%2E+RETURNN+as+a+generic+flexible+neural+toolkit+with+application+to+translation+and+speech+recognition%2E+In+Proceedings+of+ACL+2018%2C+System+Demonstrati)
58. Jonathan Shen, Patrick Nguyen, Yonghui Wu, Zhifeng Chen, Mia X Chen, Ye Jia, Anjuli Kannan, Tara Sainath, Yuan Cao, Chung-Cheng Chiu, et al. 2019. Lingvo: a modular and scalable framework for sequence-to-sequence modeling. arXiv preprint arXiv:1902.08295. — [arXiv:1902.08295](https://arxiv.org/abs/1902.08295) [find](https://scholar.google.com/scholar?q=Jonathan+Shen%2C+Patrick+Nguyen%2C+Yonghui+Wu%2C+Zhifeng+Chen%2C+Mia+X+Chen%2C+Ye+Jia%2C+Anjuli+Kannan%2C+Tara+Sainath%2C+Yuan+Cao%2C+Chung%2DCheng+Chiu%2C+et+al%2E+2019%2E+Lingvo%3A+a+modular+and+scalable+framework+for+sequence)