---
title: "Improving Automatic Speech Recognition for Non-Native English"
description: "ASR systems tuned for native (L1) English perform noticeably worse on non-native (L2) speakers. 59 references, cited by 3 papers here."
authors: ["Peter Sullivan", "Toshiko Shibano", "Muhammad Abdul-Mageed"]
year: 2022
venue: "arXiv:2202.05209"
source: "https://arxiv.org/abs/2202.05209"
doi: "10.48550/arxiv.2202.05209"
arxiv: "2202.05209"
licence: "CC BY 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/sullivan-2022-improving-automatic-speech-recognition-non/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/sullivan-2022-improving-automatic-speech-recognition-non/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2022](https://md.research.elamur.ai/free/en/papers/2022/page/5/) / Improving Automatic Speech Recognition for Non-Native English with Transfer Learning and Language Model Decoding

# Improving Automatic Speech Recognition for Non-Native English with Transfer Learning and Language Model Decoding

**Authors:** [Peter Sullivan](https://md.research.elamur.ai/free/en/authors/peter-sullivan/), [Toshiko Shibano](https://md.research.elamur.ai/free/en/authors/toshiko-shibano/) and [Muhammad Abdul-Mageed](https://md.research.elamur.ai/free/en/authors/muhammad-abdul-mageed/) · **Year:** 2022 · **Venue:** arXiv:2202.05209 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2202.05209) · [PDF](https://arxiv.org/pdf/2202.05209) · [DOI 10.48550/arxiv.2202.05209](https://doi.org/10.48550/arxiv.2202.05209) · [arXiv 2202.05209](https://arxiv.org/abs/2202.05209) · [OpenAlex](https://openalex.org/W4321513043)

**Citations:** 13 citations (OpenAlex, 2026-10-14) · 3 in our library, of the 141,840 papers whose bibliographies we read

## Summary
ASR systems tuned for native (L1) English perform noticeably worse on non-native (L2) speakers. Accents shift pronunciation, and little annotated accented audio exists for training. The authors start from the self-trained wav2vec 2.0 Large model (Wav2Vec 2.0-ST) and fine-tune it with CTC under many data conditions: L1 speech, pooled L2 accents, single L2 accents, and multi-accent zero-shot setups. L2-ARCTIC is the central L2 corpus. They compare plain best-path decoding with beam search guided by an external KenLM language model. They also study error types through Levenshtein edit operations.

Fine-tuning and LM decoding each give gains of roughly similar size, and combining them closes much of the L1–L2 gap. A key observation concerns the language knowledge the self-trained model absorbs during training. That knowledge is enough for clean L1 audio but does not hold up on accented speech, so an explicit LM still helps. This is a useful warning against assuming a strong end-to-end model needs no external linguistic prior on out-of-domain input.

## Key points
- **Base model and fine-tuning:**
  - The base is Wav2Vec 2.0-ST (LV-60). It was pre-trained on 960 h of LibriSpeech, then self-trained on 53.2k h of Libri-Light using pseudo-labels produced with LibriSpeech language models.
  - Fine-tuning uses CTC with the CNN feature extractor frozen.
  - Masking and learning-rate settings, which imitate the tri-state schedule, are chosen per Dev set.
- **Accent findings:**
  - Single-accent fine-tuning gives the largest gains on its target accent.
  - Multi-accent fine-tuning generalizes more widely.
  - Fine-tuning on domain-specific L1 speech actually hurts L2 performance.
- **Error reduction on the L2-ARCTIC dev set, relative to baseline B1** (best path decoding, no LM):

  | Model | Substitutions | Deletions | Insertions |
  |---|---|---|---|
  | Fine-tuned M4 | −43% | −55% | −30% |
  | M4 + LM | −64% | −65% | −52% |

- **LM without fine-tuning:** adding the LM to the non-fine-tuned B1 lowers WER but increases deletions.
- **Single-edit errors made by B1:**
  - 15 of 18 deletions are spelling errors.
  - All 18 substitutions are spelling errors.
  - 7 of 11 insertions are spacing errors, e.g. "to morrow".
  - The LM is especially good at fixing word spacing.

## Interesting ideas
- Isolate the contribution of two interventions with a full ±fine-tuning × ±LM grid, so gains from adaptation and from an external prior are measured separately rather than lumped together.
- Use cheap automatic edit-operation counts as an error taxonomy, then hand-inspect only the simplest single-edit cases to name the underlying linguistic cause (spelling, spacing, plural, tense).
- Do not assume knowledge a model internalized on clean in-domain data transfers to shifted input; test that explicitly, since the paper's LM-free strength vanished on accented speech.
- Contrast training on one target group, a pooled set of groups, and a held-out group to see the trade-off between peak accuracy and generalization.
- Freeze most weights (adapters, LayerNorm/attention-only tuning) when adapting, to avoid erasing knowledge the base model already holds.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): Run the same 2×2 ablation for parser retraining vs. self-induced repair rules, so we know which one actually fixes which errors.
- **I2** (tokens): Categorize tokenizer, lemma and morphology mismatches by edit type, then manually label a small single-edit subset to find systematic rule gaps.
- **I3** (tree): Evaluate gparse on learner or non-standard English separately, and keep explicit lexical and grammatical constraints available as a fallback when the shift appears.
- **I4** (learn): Train silver-data parsers per genre, pooled, and with one genre held out, to choose between specialist and generalist models.
- **I5** (tree): When adapting gparse to a new UD dialect or treebank, tune only small adapter parts so behaviour on the original treebanks is preserved.

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

## Institutions

- [University of British Columbia](https://md.research.elamur.ai/free/en/universities/university-of-british-columbia/) — University, Vancouver, Canada · 46 works · 33,673 citations

📚 Shelves (7): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 798 works · [💭 Experimental and Cognitive Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/experimental-and-cognitive-psychology/) · 463 works · [🔊 Signal Processing](https://md.research.elamur.ai/free/en/shelves/computer-science/signal-processing/) · 213 works

🧭 Topics (3): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 1,343 works · [English](https://md.research.elamur.ai/free/en/topics/english/) · 1,373 works · [🎙️ Speech & spoken language](https://md.research.elamur.ai/free/en/topics/speech/) · 233 works

## Cite

Cite as: Peter Sullivan, Toshiko Shibano and Muhammad Abdul-Mageed. 2022. Improving Automatic Speech Recognition for Non-Native English with Transfer Learning and Language Model Decoding. arXiv:2202.05209.

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

```bibtex
@misc{sullivan2022improving,
  title = {{Improving Automatic Speech Recognition for Non-Native English with Transfer Learning and Language Model Decoding}},
  author = {Peter Sullivan, Toshiko Shibano and Muhammad Abdul-Mageed},
  year = {2022},
  doi = {10.48550/arxiv.2202.05209},
  eprint = {2202.05209},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2202.05209}
}
```

- [← previous Controllable Lexical Simplification for English](https://md.research.elamur.ai/free/en/papers/sheang-2022-controllable-lexical-simplification-english/)
- [next → Label distributions help implicit discourse relation classification](https://md.research.elamur.ai/free/en/papers/yung-2022-label-distributions-help-implicit-discourse/)
- [↑ back to papers of 2022](https://md.research.elamur.ai/free/en/papers/2022/page/5/)

## References

59 references, 10 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Baevski A, Schneider S, Auli M (2019) vq-wav2vec: Self-supervised learning of discrete speech representations. arXiv preprint arXiv:191005453 [find](https://scholar.google.com/scholar?q=Baevski+A%2C+Schneider+S%2C+Auli+M+%282019%29+vq%2Dwav2vec%3A+Self%2Dsupervised+learning+of+discrete+speech+representations%2E+arXiv+preprint+arXiv%3A191005453)
2. Baevski A, Zhou H, Mohamed A, Auli M (2020) [wav2vec 2.0: A framework for self-supervised learning of speech representations](https://md.research.elamur.ai/free/en/papers/baevski-2020-wav2vec-2-0-framework-self/). arXiv preprint arXiv:200611477
3. Bearman A, Josund K, Fiore G (2017) Accent conversion using artificial neural networks. Tech. rep., Stanford University, Tech. Rep, Tech. Rep [find](https://scholar.google.com/scholar?q=Bearman+A%2C+Josund+K%2C+Fiore+G+%282017%29+Accent+conversion+using+artificial+neural+networks%2E+Tech%2E+rep%2E%2C+Stanford+University%2C+Tech%2E+Rep%2C+Tech%2E+Rep)
4. Chan W, Jaitly N, Le Q, Vinyals O (2016) Listen, attend and spell: A neural network for large vocabulary conversational speech recognition. In: 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 4960–4964 [find](https://scholar.google.com/scholar?q=Chan+W%2C+Jaitly+N%2C+Le+Q%2C+Vinyals+O+%282016%29+Listen%2C+attend+and+spell%3A+A+neural+network+for+large+vocabulary+conversational+speech+recognition%2E+In%3A+2016+IEEE+International+Conference+on+Acoustics%2C+Speech+)
5. Chorowski JK, Bahdanau D, Serdyuk D, Cho K, Bengio Y (2015) [Attention-based models for speech recognition](https://md.research.elamur.ai/free/en/papers/chorowski-2015-attention-based-models-speech-recognition/). In: Advances in neural information processing systems, pp 577–585 [↗](https://doi.org/10.48550/arxiv.1506.07503)
6. Chung YA, Hsu WN, Tang H, Glass J (2019) An unsupervised autoregressive model for speech representation learning. arXiv preprint arXiv:190403240 [find](https://scholar.google.com/scholar?q=Chung+YA%2C+Hsu+WN%2C+Tang+H%2C+Glass+J+%282019%29+An+unsupervised+autoregressive+model+for+speech+representation+learning%2E+arXiv+preprint+arXiv%3A190403240)
7. Crystal D (2003) English as a global language. Ernst Klett Sprachen [↗](https://doi.org/10.1017/cbo9780511486999)
8. Das N, Bodapati S, Sunkara M, Srinivasan S, Chau DH (2021) Best of both worlds: Robust accented speech recognition with adversarial transfer learning. arXiv preprint arXiv:210305834 [find](https://scholar.google.com/scholar?q=Das+N%2C+Bodapati+S%2C+Sunkara+M%2C+Srinivasan+S%2C+Chau+DH+%282021%29+Best+of+both+worlds%3A+Robust+accented+speech+recognition+with+adversarial+transfer+learning%2E+arXiv+preprint+arXiv%3A210305834)
9. Devlin J, Chang MW, Lee K, Toutanova K (2018) [Bert: Pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). arXiv preprint arXiv:181004805
10. Futami H, Inaguma H, Ueno S, Mimura M, Sakai S, Kawahara T (2020) Distilling the knowledge of bert for sequence-to-sequence asr. arXiv preprint arXiv:200803822 [find](https://scholar.google.com/scholar?q=Futami+H%2C+Inaguma+H%2C+Ueno+S%2C+Mimura+M%2C+Sakai+S%2C+Kawahara+T+%282020%29+Distilling+the+knowledge+of+bert+for+sequence%2Dto%2Dsequence+asr%2E+arXiv+preprint+arXiv%3A200803822)
11. Graves A (2012) Connectionist temporal classification. In: Supervised Sequence Labelling with Recurrent Neural Networks, Springer, pp 61–93 [↗](https://doi.org/10.1007/978-3-642-24797-2_7)
12. Graves A, Jaitly N (2014) Towards end-to-end speech recognition with recurrent neural networks. In: International conference on machine learning, pp 1764– 1772 [find](https://scholar.google.com/scholar?q=Graves+A%2C+Jaitly+N+%282014%29+Towards+end%2Dto%2Dend+speech+recognition+with+recurrent+neural+networks%2E+In%3A+International+conference+on+machine+learning%2C+pp+1764%E2%80%93+1772)
13. Graves A, Fernández S, Gomez F, Schmidhuber J (2006) Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks. In: Proceedings of the 23rd international conference on Machine learning, pp 369–376 [find](https://scholar.google.com/scholar?q=Graves+A%2C+Fern%C3%A1ndez+S%2C+Gomez+F%2C+Schmidhuber+J+%282006%29+Connectionist+temporal+classification%3A+labelling+unsegmented+sequence+data+with+recurrent+neural+networks%2E+In%3A+Proceedings+of+the+23rd+internationa)
14. Gulati A, Qin J, Chiu CC, Parmar N, Zhang Y, Yu J, Han W, Wang S, Zhang Z, Wu Y, et al. (2020) [Conformer: Convolution-augmented transformer for speech recognition](https://md.research.elamur.ai/free/en/papers/gulati-2020-conformer-convolution-augmented-transformer-speech/). arXiv preprint arXiv:200508100
15. Guliani D, Beaufays F, Motta G (2021) Training speech recognition models with federated learning: A quality/cost framework. In: ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 3080–3084 [find](https://scholar.google.com/scholar?q=Guliani+D%2C+Beaufays+F%2C+Motta+G+%282021%29+Training+speech+recognition+models+with+federated+learning%3A+A+quality%2Fcost+framework%2E+In%3A+ICASSP+2021%2D2021+IEEE+International+Conference+on+Acoustics%2C+Speech+and+)
16. Hannun AY, Maas AL, Jurafsky D, Ng AY (2014) First-pass large vocabulary continuous speech recognition using bi-directional recurrent dnns. arXiv preprint arXiv:14082873 [find](https://scholar.google.com/scholar?q=Hannun+AY%2C+Maas+AL%2C+Jurafsky+D%2C+Ng+AY+%282014%29+First%2Dpass+large+vocabulary+continuous+speech+recognition+using+bi%2Ddirectional+recurrent+dnns%2E+arXiv+preprint+arXiv%3A14082873)
17. Heafield K (2011) [Kenlm: Faster and smaller language model queries](https://md.research.elamur.ai/free/en/papers/heafield-2011-kenlm-faster-smaller-language-model/). In: Proceedings of the sixth workshop on statistical machine translation, pp 187– 197
18. Hori T, Cho J, Watanabe S (2018) End-to-end speech recognition with word-based rnn language models. In: 2018 IEEE Spoken Language Technology Workshop (SLT), IEEE, pp 389–396 [find](https://scholar.google.com/scholar?q=Hori+T%2C+Cho+J%2C+Watanabe+S+%282018%29+End%2Dto%2Dend+speech+recognition+with+word%2Dbased+rnn+language+models%2E+In%3A+2018+IEEE+Spoken+Language+Technology+Workshop+%28SLT%29%2C+IEEE%2C+pp+389%E2%80%93396)
19. Hou J, Guo P, Sun S, Soong FK, Hu W, Xie L (2019) Domain adversarial training for improving keyword spotting performance of esl speech. In: ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 8122–8126 [find](https://scholar.google.com/scholar?q=Hou+J%2C+Guo+P%2C+Sun+S%2C+Soong+FK%2C+Hu+W%2C+Xie+L+%282019%29+Domain+adversarial+training+for+improving+keyword+spotting+performance+of+esl+speech%2E+In%3A+ICASSP+2019%2D2019+IEEE+International+Conference+on+Acoustics%2C)
20. Houlsby N, Giurgiu A, Jastrzebski S, Morrone B, De Laroussilhe Q, Gesmundo A, Attariyan M, Gelly S (2019) Parameter-efficient transfer learning for nlp. In: International Conference on Machine Learning, PMLR, pp 2790–2799 [↗](https://doi.org/10.48550/arxiv.1902.00751)
21. Hu H, Yang X, Raeesy Z, Guo J, Keskin G, Arsikere H, Rastrow A, Stolcke A, Maas R (2021) Redat: Accent-invariant representation for end-to-end asr by domain adversarial training with relabeling. In: ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 6408–6412 [↗](https://doi.org/10.48550/arxiv.2012.07353)
22. Hwang K, Sung W (2016) Character-level incremental speech recognition with recurrent neural networks. In: 2016 IEEE international conference on acoustics, speech and signal processing (ICASSP), IEEE, pp 5335–5339 [find](https://scholar.google.com/scholar?q=Hwang+K%2C+Sung+W+%282016%29+Character%2Dlevel+incremental+speech+recognition+with+recurrent+neural+networks%2E+In%3A+2016+IEEE+international+conference+on+acoustics%2C+speech+and+signal+processing+%28ICASSP%29%2C+IEEE%2C+)
23. Jain A, Upreti M, Jyothi P (2018) Improved accented speech recognition using accent embeddings and multi-task learning. In: Interspeech, pp 2454–2458 [find](https://scholar.google.com/scholar?q=Jain+A%2C+Upreti+M%2C+Jyothi+P+%282018%29+Improved+accented+speech+recognition+using+accent+embeddings+and+multi%2Dtask+learning%2E+In%3A+Interspeech%2C+pp+2454%E2%80%932458)
24. Kahn J, Rivière M, Zheng W, Kharitonov E, Xu Q, Mazaré PE, Karadayi J, Liptchinsky V, Collobert R, Fuegen C, et al. (2020) Libri-light: A benchmark for asr with limited or no supervision. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 7669–7673 [find](https://scholar.google.com/scholar?q=Kahn+J%2C+Rivi%C3%A8re+M%2C+Zheng+W%2C+Kharitonov+E%2C+Xu+Q%2C+Mazar%C3%A9+PE%2C+Karadayi+J%2C+Liptchinsky+V%2C+Collobert+R%2C+Fuegen+C%2C+et+al%2E+%282020%29+Libri%2Dlight%3A+A+benchmark+for+asr+with+limited+or+no+supervision%2E+In%3A+ICASSP+2)
25. Kominek J, Black AW, Ver V (2003) Cmu arctic databases for speech synthesis [find](https://scholar.google.com/scholar?q=Kominek+J%2C+Black+AW%2C+Ver+V+%282003%29+Cmu+arctic+databases+for+speech+synthesis)
26. Kunze J, Kirsch L, Kurenkov I, Krug A, Johannsmeier J, Stober S (2017) Transfer learning for speech recognition on a budget. arXiv preprint arXiv:170600290 [find](https://scholar.google.com/scholar?q=Kunze+J%2C+Kirsch+L%2C+Kurenkov+I%2C+Krug+A%2C+Johannsmeier+J%2C+Stober+S+%282017%29+Transfer+learning+for+speech+recognition+on+a+budget%2E+arXiv+preprint+arXiv%3A170600290)
27. Li X, Wang C, Tang Y, Tran C, Tang Y, Pino J, Baevski A, Conneau A, Auli M (2020) Multilingual speech translation with efficient finetuning of pretrained models. arXiv preprint arXiv:201012829 [find](https://scholar.google.com/scholar?q=Li+X%2C+Wang+C%2C+Tang+Y%2C+Tran+C%2C+Tang+Y%2C+Pino+J%2C+Baevski+A%2C+Conneau+A%2C+Auli+M+%282020%29+Multilingual+speech+translation+with+efficient+finetuning+of+pretrained+models%2E+arXiv+preprint+arXiv%3A201012829)
28. Ling S, Liu Y, Salazar J, Kirchhoff K (2020) Deep contextualized acoustic representations for semi-supervised speech recognition. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 6429–6433 [find](https://scholar.google.com/scholar?q=Ling+S%2C+Liu+Y%2C+Salazar+J%2C+Kirchhoff+K+%282020%29+Deep+contextualized+acoustic+representations+for+semi%2Dsupervised+speech+recognition%2E+In%3A+ICASSP+2020%2D2020+IEEE+International+Conference+on+Acoustics%2C+Speec)
29. Liu AT, Yang Sw, Chi PH, Hsu Pc, Lee Hy (2020) Mockingjay: Unsupervised speech representation learning with deep bidirectional transformer encoders. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 6419–6423 [find](https://scholar.google.com/scholar?q=Liu+AT%2C+Yang+Sw%2C+Chi+PH%2C+Hsu+Pc%2C+Lee+Hy+%282020%29+Mockingjay%3A+Unsupervised+speech+representation+learning+with+deep+bidirectional+transformer+encoders%2E+In%3A+ICASSP+2020%2D2020+IEEE+International+Conference+)
30. Liu S, Wang D, Cao Y, Sun L, Wu X, Kang S, Wu Z, Liu X, Su D, Yu D, et al. (2020) End-to-end accent conversion without using native utterances. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 6289–6293 [find](https://scholar.google.com/scholar?q=Liu+S%2C+Wang+D%2C+Cao+Y%2C+Sun+L%2C+Wu+X%2C+Kang+S%2C+Wu+Z%2C+Liu+X%2C+Su+D%2C+Yu+D%2C+et+al%2E+%282020%29+End%2Dto%2Dend+accent+conversion+without+using+native+utterances%2E+In%3A+ICASSP+2020%2D2020+IEEE+International+Conference+on+Ac)
31. Livescu K, Glass J (2000) Lexical modeling of non-native speech for automatic speech recognition. In: 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No. 00CH37100), IEEE, vol 3, pp 1683–1686 [find](https://scholar.google.com/scholar?q=Livescu+K%2C+Glass+J+%282000%29+Lexical+modeling+of+non%2Dnative+speech+for+automatic+speech+recognition%2E+In%3A+2000+IEEE+International+Conference+on+Acoustics%2C+Speech%2C+and+Signal+Processing%2E+Proceedings+%28Cat%2E+)
32. Lowerre BT (1976) The harpy speech recognition system. Carnegie Mellon University [find](https://scholar.google.com/scholar?q=Lowerre+BT+%281976%29+The+harpy+speech+recognition+system%2E+Carnegie+Mellon+University)
33. Maas A, Xie Z, Jurafsky D, Ng AY (2015) Lexicon-free conversational speech recognition with neural networks. In: Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp 345–354 [find](https://scholar.google.com/scholar?q=Maas+A%2C+Xie+Z%2C+Jurafsky+D%2C+Ng+AY+%282015%29+Lexicon%2Dfree+conversational+speech+recognition+with+neural+networks%2E+In%3A+Proceedings+of+the+2015+Conference+of+the+North+American+Chapter+of+the+Association+for)
34. Matassoni M, Gretter R, Falavigna D, Giuliani D (2018) Non-native children speech recognition through transfer learning. In: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 6229–6233 [find](https://scholar.google.com/scholar?q=Matassoni+M%2C+Gretter+R%2C+Falavigna+D%2C+Giuliani+D+%282018%29+Non%2Dnative+children+speech+recognition+through+transfer+learning%2E+In%3A+2018+IEEE+International+Conference+on+Acoustics%2C+Speech+and+Signal+Processi)
35. Meister C, Vieira T, Cotterell R (2020) [If beam search is the answer, what was the question?](https://md.research.elamur.ai/free/en/papers/meister-2020-if-beam-search-is-answer/) arXiv preprint arXiv:201002650
36. Miao Y, Gowayyed M, Metze F (2015) Eesen: End-to-end speech recognition using deep rnn models and wfst-based decoding. In: 2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU), IEEE, pp 167–174 [find](https://scholar.google.com/scholar?q=Miao+Y%2C+Gowayyed+M%2C+Metze+F+%282015%29+Eesen%3A+End%2Dto%2Dend+speech+recognition+using+deep+rnn+models+and+wfst%2Dbased+decoding%2E+In%3A+2015+IEEE+Workshop+on+Automatic+Speech+Recognition+and+Understanding+%28ASRU%29%2C+)
37. Oord Avd, Li Y, Vinyals O (2018) [Representation learning with contrastive predictive coding](https://md.research.elamur.ai/free/en/papers/oord-2018-representation-learning-contrastive-predictive-coding/). arXiv preprint arXiv:180703748
38. Pan SJ, Yang Q (2009) A survey on transfer learning. IEEE Transactions on knowledge and data engineering 22(10):1345–1359 [↗](https://doi.org/10.1109/tkde.2009.191)
39. Panayotov V, Chen G, Povey D, Khudanpur S (2015) Librispeech: an asr corpus based on public domain audio books. In: 2015 IEEE international conference on acoustics, speech and signal processing (ICASSP), IEEE, pp 5206–5210 [find](https://scholar.google.com/scholar?q=Panayotov+V%2C+Chen+G%2C+Povey+D%2C+Khudanpur+S+%282015%29+Librispeech%3A+an+asr+corpus+based+on+public+domain+audio+books%2E+In%3A+2015+IEEE+international+conference+on+acoustics%2C+speech+and+signal+processing+%28ICASS)
40. Park DS, Chan W, Zhang Y, Chiu CC, Zoph B, Cubuk ED, Le QV (2019) [Specaugment: A simple data augmentation method for automatic speech recognition](https://md.research.elamur.ai/free/en/papers/park-2019-specaugment-simple-data-augmentation-method/). arXiv preprint arXiv:190408779
41. Ping TT (2008) Automatic speech recognition for non-native speakers. PhD thesis, Université Joseph-Fourier-Grenoble I [find](https://scholar.google.com/scholar?q=Ping+TT+%282008%29+Automatic+speech+recognition+for+non%2Dnative+speakers%2E+PhD+thesis%2C+Universit%C3%A9+Joseph%2DFourier%2DGrenoble+I)
42. Radzikowski K, Wang L, Yoshie O, Nowak R (2021) Accent modification for speech recognition of non-native speakers using neural style transfer. EURASIP Journal on Audio, Speech, and Music Processing 2021(1):1–10 [↗](https://doi.org/10.1186/s13636-021-00199-3)
43. Sak H, Senior A, Rao K, Irsoy O, Graves A, Beaufays F, Schalkwyk J (2015) Learning acoustic frame labeling for speech recognition with recurrent neural networks. In: 2015 IEEE international conference on acoustics, speech and signal processing (ICASSP), IEEE, pp 4280–4284 [find](https://scholar.google.com/scholar?q=Sak+H%2C+Senior+A%2C+Rao+K%2C+Irsoy+O%2C+Graves+A%2C+Beaufays+F%2C+Schalkwyk+J+%282015%29+Learning+acoustic+frame+labeling+for+speech+recognition+with+recurrent+neural+networks%2E+In%3A+2015+IEEE+international+conference)
44. Schneider S, Baevski A, Collobert R, Auli M (2019) wav2vec: Unsupervised pre-training for speech recognition. arXiv preprint arXiv:190405862 [find](https://scholar.google.com/scholar?q=Schneider+S%2C+Baevski+A%2C+Collobert+R%2C+Auli+M+%282019%29+wav2vec%3A+Unsupervised+pre%2Dtraining+for+speech+recognition%2E+arXiv+preprint+arXiv%3A190405862)
45. Shi X, Yu F, Lu Y, Liang Y, Feng Q, Wang D, Qian Y, Xie L (2021) [The accented english speech recognition challenge 2020: open datasets, tracks, baselines, results and methods](https://md.research.elamur.ai/free/en/papers/shi-2021-accented-english-speech-recognition-challenge/). In: ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 6918–6922 [↗](https://doi.org/10.48550/arxiv.2102.10233)
46. Shibano T, Zhang X, Li MT, Cho H, Sullivan P, Abdul-Mageed M (2021) [Speech technology for everyone: Automatic speech recognition for non-native english with transfer learning](https://md.research.elamur.ai/free/en/papers/shibano-2021-speech-technology-everyone-automatic-speech-recognition/). arXiv preprint arXiv:211000678
47. Sun S, Yeh CF, Hwang MY, Ostendorf M, Xie L (2018) Domain adversarial training for accented speech recognition. In: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 4854– 4858 [find](https://scholar.google.com/scholar?q=Sun+S%2C+Yeh+CF%2C+Hwang+MY%2C+Ostendorf+M%2C+Xie+L+%282018%29+Domain+adversarial+training+for+accented+speech+recognition%2E+In%3A+2018+IEEE+International+Conference+on+Acoustics%2C+Speech+and+Signal+Processing+%28ICASS)
48. Synnaeve G, Xu Q, Kahn J, Likhomanenko T, Grave E, Pratap V, Sriram A, Liptchinsky V, Collobert R (2019) End-to-end asr: from supervised to semi-supervised learning with modern architectures. arXiv preprint arXiv:191108460 [find](https://scholar.google.com/scholar?q=Synnaeve+G%2C+Xu+Q%2C+Kahn+J%2C+Likhomanenko+T%2C+Grave+E%2C+Pratap+V%2C+Sriram+A%2C+Liptchinsky+V%2C+Collobert+R+%282019%29+End%2Dto%2Dend+asr%3A+from+supervised+to+semi%2Dsupervised+learning+with+modern+architectures%2E+arXiv+pr)
49. Turan MAT, Vincent E, Jouvet D (2020) Achieving multi-accent asr via unsupervised acoustic model adaptation. In: INTERSPEECH 2020 [find](https://scholar.google.com/scholar?q=Turan+MAT%2C+Vincent+E%2C+Jouvet+D+%282020%29+Achieving+multi%2Daccent+asr+via+unsupervised+acoustic+model+adaptation%2E+In%3A+INTERSPEECH+2020)
50. Viglino T, Motlicek P, Cernak M (2019) End-to-end accented speech recognition. In: Interspeech, pp 2140–2144 [find](https://scholar.google.com/scholar?q=Viglino+T%2C+Motlicek+P%2C+Cernak+M+%282019%29+End%2Dto%2Dend+accented+speech+recognition%2E+In%3A+Interspeech%2C+pp+2140%E2%80%932144)
51. Wang D, Zheng TF (2015) Transfer learning for speech and language processing. In: 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), IEEE, pp 1225–1237 [find](https://scholar.google.com/scholar?q=Wang+D%2C+Zheng+TF+%282015%29+Transfer+learning+for+speech+and+language+processing%2E+In%3A+2015+Asia%2DPacific+Signal+and+Information+Processing+Association+Annual+Summit+and+Conference+%28APSIPA%29%2C+IEEE%2C+pp+1225%E2%80%931)
52. Wang Y, Luan H, Yuan J, Wang B, Lin H (2020) Laix corpus of chinese learner english: Towards a benchmark for l2 english asr. In: INTERSPEECH, pp 414–418 [find](https://scholar.google.com/scholar?q=Wang+Y%2C+Luan+H%2C+Yuan+J%2C+Wang+B%2C+Lin+H+%282020%29+Laix+corpus+of+chinese+learner+english%3A+Towards+a+benchmark+for+l2+english+asr%2E+In%3A+INTERSPEECH%2C+pp+414%E2%80%93418)
53. Wang Y, Mohamed A, Le D, Liu C, Xiao A, Mahadeokar J, Huang H, Tjandra A, Zhang X, Zhang F, et al. (2020) Transformer-based acoustic modeling for hybrid speech recognition. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 6874–6878 [find](https://scholar.google.com/scholar?q=Wang+Y%2C+Mohamed+A%2C+Le+D%2C+Liu+C%2C+Xiao+A%2C+Mahadeokar+J%2C+Huang+H%2C+Tjandra+A%2C+Zhang+X%2C+Zhang+F%2C+et+al%2E+%282020%29+Transformer%2Dbased+acoustic+modeling+for+hybrid+speech+recognition%2E+In%3A+ICASSP+2020%2D2020+IEEE+I)
54. Wang Z, Schultz T, Waibel A (2003) Comparison of acoustic model adaptation techniques on non-native speech. In: 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings.(ICASSP’03)., IEEE, vol 1, pp I–I [find](https://scholar.google.com/scholar?q=Wang+Z%2C+Schultz+T%2C+Waibel+A+%282003%29+Comparison+of+acoustic+model+adaptation+techniques+on+non%2Dnative+speech%2E+In%3A+2003+IEEE+International+Conference+on+Acoustics%2C+Speech%2C+and+Signal+Processing%2C+2003%2E+Pr)
55. Watanabe S, Hori T, Kim S, Hershey JR, Hayashi T (2017) Hybrid ctc/attention architecture for end-to-end speech recognition. IEEE Journal of Selected Topics in Signal Processing 11(8):1240–1253 [↗](https://doi.org/10.1109/jstsp.2017.2763455)
56. Xu Q, Baevski A, Likhomanenko T, Tomasello P, Conneau A, Collobert R, Synnaeve G, Auli M (2021) Self-training and pre-training are complementary for speech recognition. In: ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 3030–3034 [↗](https://doi.org/10.48550/arxiv.2010.11430)
57. Yu W, Freiwald J, Tewes S, Huennemeyer F, Kolossa D (2021) Federated learning in asr: Not as easy as you think. In: Speech Communication; 14th ITG Conference, VDE, pp 1–5 [↗](https://doi.org/10.48550/arxiv.2109.15108)
58. Zenkel T, Sanabria R, Metze F, Niehues J, Sperber M, Stüker S, Waibel A (2017) Comparison of decoding strategies for ctc acoustic models. arXiv preprint arXiv:170804469 [find](https://scholar.google.com/scholar?q=Zenkel+T%2C+Sanabria+R%2C+Metze+F%2C+Niehues+J%2C+Sperber+M%2C+St%C3%BCker+S%2C+Waibel+A+%282017%29+Comparison+of+decoding+strategies+for+ctc+acoustic+models%2E+arXiv+preprint+arXiv%3A170804469)
59. Zhao G, Sonsaat S, Silpachai AO, Lucic I, Chukharev-Hudilainen E, Levis J, Gutierrez-Osuna R (2018) L2-arctic: A non-native english speech corpus. Perception Sensing Instrumentation Lab [find](https://scholar.google.com/scholar?q=Zhao+G%2C+Sonsaat+S%2C+Silpachai+AO%2C+Lucic+I%2C+Chukharev%2DHudilainen+E%2C+Levis+J%2C+Gutierrez%2DOsuna+R+%282018%29+L2%2Darctic%3A+A+non%2Dnative+english+speech+corpus%2E+Perception+Sensing+Instrumentation+Lab)

*Updated 24 October 2026*