Pith. sign in

REVIEW 1 major objections 7 minor 40 references

PHRASED: Phrase Dictionary Biasing for Speech Translation

T0 review · 1 major / 7 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read By matching source phrases in an intermediate transcript before adding target-phrase bonuses, PHRASED improves phrase recall by 21% over phrase list biasing on streaming speech translation and by 85% for a multimodal LLM.

desk verdict PHRASED is a practical, honest biasing recipe for streaming ST and LLM ST, but its headline gains are gated on an unmeasured prerequisite: the source phrase must appear verbatim in the ASR/CTC text. read the letter →

arxiv 2506.09175 v1 pith:QZNFDL3A submitted 2025-06-10 cs.CL cs.AIcs.SDeess.AS

Peidong Wang , Jian Xue , Rui Zhao , Junkun Chen , Aswin Shanmugam Subramanian , Jinyu Li This is my paper
classification cs.CLcs.AIcs.SDeess.AS
keywords speechtranslationphrasedictionarybiasingcontextualstreamingmultimodallargelanguagemodelshallowfusionrecallCTCcompression
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

PHRASED is a decoding-time method for fixing rare-phrase translation in speech translation without fine-tuning. It uses a phrase dictionary of source-language/target-language pairs and adds a bonus to a target phrase during decoding only if its source phrase appears in an intermediate text representation of the audio. On the streaming transducer model CTC-GMM, this selection-based biasing reaches 52.25% phrase recall versus 43.24% for phrase list biasing, a 21% relative improvement, with BLEU rising from 19.9 to 20.0. On the multimodal language model Phi-4-multimodal, the same idea turns external phrase information into usable prompts and lifts phrase recall from 36.04% to as high as 66.67%, an 85% relative gain. The paper argues that the gain comes from knowing which target phrases are actually relevant, not from applying a larger bonus.

What carries the argument

The load-bearing object is the phrase dictionary $\{I:O\}$, pairs of source-language phrases $I$ and target-language phrases $O$, combined with a gated scoring rule. PHRASED interpolates the model score with a bonus from the selected target phrases: $y^* = \arg\max_y \log P(y|x) + \lambda \log P_O(y) + \mu \log P_{O^m}(y|z)$, where $z$ is the intermediate representation and $O^m$ is the subset of target phrases whose source side matched $z$. The mechanism that carries the argument is the gate: at each decoding step the source phrases are matched verbatim against $z$, and bonuses are added only for the matched target phrases, with a bookkeeping step that removes partially matched bonuses when a phrase does not complete. This gate is what distinguishes PHRASED from phrase list biasing, and it is the component the ablation isolates when the large unconditional bonus fails.

What would settle it

Compute phrase recall on test utterances with rare source phrases, separating cases where the phrase does and does not appear verbatim in the CTC/ASR intermediate output. The mechanism predicts that PHRASED's recall gain is concentrated entirely in the matched subset and that the unmatched subset stays at the unassisted baseline; if recall improves on unmatched utterances, then the stated selection gate is not what produces the gain.

Watch

Extended reading notes

Core claim

The central claim is that the missing ingredient in speech-translation biasing is the link between what was said and what should be output. Existing phrase list biasing boosts every target phrase on a list regardless of the audio; PHRASED instead pairs each target phrase with its source-language counterpart and consults an intermediate representation z of the utterance (ASR text; for CTC-GMM, the CTC output converted from byte-level tokens to text at each decoding step). Target phrases whose source side appears in z receive a bonus; others do not. In the paper's experiments, the gated version beats the ungated one even when the ungated bonus is made equally large: PLB with a 4.0 bonus scores 4.6 BLEU and 49.55% recall, while PHRASED with selection scores 20.0 BLEU and 52.25% recall on the same 3K phrase list. For the multimodal LLM, phrase-list prompting makes the model refuse the task entirely, while PHRASED's phrase selection and joint biasing reach 54.95% and 66.67% recall, respectively. The discovery is that selection, not bonus strength, is what makes phrase biasing work for speech translation.

Load-bearing premise

The method only works when the source phrase is already present, word for word, in the intermediate transcript: if the speech recognizer or CTC output misses the phrase, no match occurs, no bonus is added, and decoding falls back to the unassisted model.

Editorial extensions

If this is right

  • Adding or updating phrases becomes an inference-time operation: users can supply a new dictionary entry and see the phrase used without retraining or storing phrase information in model weights.
  • Bonus size is not the driver: an equally large but ungated bonus drops BLEU to 4.6, while gated selection holds BLEU at 20.0, so future biasing systems should focus on when to apply a bonus.
  • Multimodal LLMs can be made to use external phrase knowledge through prompts, avoiding the total failure that a several-thousand-phrase list prompt causes.
  • Because the selection mechanism works on both a streaming transducer model and a multimodal LLM, it is a general strategy rather than a fix for one architecture.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The method's ceiling is set by the recognizability of source phrases in the intermediate representation, so its headroom is tied to the quality of the CTC/ASR module; pairing PHRASED with audio embeddings or a stronger recognizer is the natural next step.
  • The two-pass design for the multimodal LLM (ASR pass, then translation pass) roughly doubles decoding cost, so a single-pass variant that interleaves phrase matching with generation would be needed before the approach suits low-latency streaming use.
  • The same 'match on a source-side hypothesis, then bias the target' principle should transfer to other generation tasks where a verifiable intermediate exists, such as ASR contextual biasing with a mapped rewrite, or cascaded ASR plus machine translation.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 7 minor

Summary. The paper proposes PHRASED, a phrase dictionary biasing method for speech translation that uses source-target phrase pairs. For streaming transducer models (CTC-GMM), it matches source phrases in an intermediate ASR/CTC text representation and adds a decoding bonus to the corresponding target phrases; for multimodal LLMs (Phi-4-multimodal), it inserts the selected and/or paired phrases into the prompt. Experiments on Chinese-English RealSI report that PHRASED phrase selection improves phrase recall by 21% relatively over phrase list biasing for CTC-GMM (Table 2), and that joint biasing improves recall by 85% relatively over the LLM baseline (Table 5). The paper also includes an oracle phrase-list analysis (Table 3) and a control showing that a larger PLB bonus does not reproduce the PHRASED recall gain.

Significance. If the claimed gains hold, PHRASED is a simple, decoding-time method to improve rare-phrase translation without fine-tuning, applicable to both streaming transducer-based ST and multimodal LLMs. The manuscript is commendably transparent: it reports an explicit large-bonus PLB ablation (Table 2) that isolates phrase selection from the extra bonus, an oracle dataset-wise phrase list analysis (Table 3), and a frank limitations section. The main uncertainty is not circularity but a missing measurement: the selection step can only act when the source phrase appears verbatim in the intermediate representation, and the paper does not report how often that prerequisite holds. This gap is load-bearing for interpreting the headline numbers.

major comments (1)
  1. [§3.2.1, §3.3, §3.4] The selection step is gated on the source-language phrase appearing verbatim in z, which is the CTC output for CTC-GMM and a separate ASR pass for Phi-4-multimodal. The paper reports no measurement of how often the phrases in I actually occur in z on RealSI, nor an oracle-z ablation in which z is replaced by the reference transcript. Since the motivating scenario is rare phrases, which are exactly the phrases that ASR/CTC models tend to misrecognize, the reported phrase recalls (e.g., 52.25% in Table 2 and 66.67% in Table 5) are upper bounds conditioned on a prerequisite whose rate is unknown. Please add (i) the match rate of I in z on the test set, (ii) phrase recall conditioned on whether the phrase was matched, and (iii) an oracle-z comparison to quantify the headroom. This does not invalidate the method, but it is necessary to interpret the headline improvements.
minor comments (7)
  1. [§3.3] The text says 'Figure 1 explains the PHRASEDjoint_biasing method for CTC-GMM', but the CTC-GMM variant studied in this paper is phrase_selection; please correct the terminology.
  2. [§4.3] The test set is described as covering 10 domains, but only 9 are enumerated (technology, health, education, finance, law, environment, entertainment, science, art); please add the missing domain or adjust the number.
  3. [§4.1.2] The list of languages ends with 'Germany'; it should be 'German'.
  4. [§5.1, Table 2] The row 'PLB w/ large bonus' is not defined in the experimental setup; please specify that it uses the same phrase list with the total bonus raised to 4.0 (or state the exact λ value used).
  5. [§3.2.1, Eq. (4)] The quantities P_O(y) and P_{O_m}(y|z) are written as probabilities but their functional form (shallow-fusion word-piece scores) is not defined; please add formal definitions for reproducibility.
  6. [§4.3] The phrase recall metric counts each target phrase once regardless of multiplicity; given that phrases can occur multiple times in a reference, please report whether the method rankings are stable under a per-occurrence variant.
  7. [Abstract and §5.4] The '85% relative improvement' is computed against the Phi-4-multimodal baseline (36.04% to 66.67%), not against the PLB baseline; please state this explicitly in the abstract to avoid ambiguity.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity; PHRASED's gains are empirical and the only author-overlapping model citation (CTC-GMM) is not load-bearing.

full rationale

The central claims are empirical comparisons on the held-out RealSI test set. PHRASED phrase_selection (Eq. 4) augments the ST decoding score with a bonus for target phrases whose source-side match appears in the intermediate representation z, and the phrase recall metric is computed against human-annotated references rather than being defined by the selection criterion. The large-bonus ablation in Table 2 (PLB w/ large bonus: BLEU 4.6, recall 49.55%) isolates phrase selection from bonus magnitude, and Table 3 transparently labels the oracle dataset-wise list as an analysis condition rather than a hidden test-set leakage. The CTC-GMM model is cited from Zhao et al. (2024b) with overlapping authors, but it is used as a concrete streaming ST system whose CTC output supplies z; this is a system choice, not a loaded theorem, fitted parameter, or uniqueness argument that forces the reported outcome. The paper's acknowledged limitation in Section 7 that only ASR-text representations were explored affects practical applicability but does not make the derivation circular: the output is not identical to the input by construction. The paper does not document how decoding weights (lambda, mu, bonus) were tuned, but these are decoding-time weights and the bonus-size confound is directly addressed by the ablation.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The ledger is light: the paper contributes a decoding-time algorithm, not new model parameters or entities. The free parameters are the decoding weights lambda and mu, the per-token bonus (3.0 for streaming ST), and the hand-built 3K phrase list, none of which enters model weights. Load-bearing axioms are the fidelity of the CTC/ASR intermediate representation for phrase matching and the correctness of the user dictionary. Evaluation rests on a custom recall convention and on the RealSI phrase annotations. Invented entities are absent because the phrase dictionary is external input, not a new model component.

free parameters (3)
  • Phrase list biasing weight lambda = 1.0 (streaming ST); 0 for LLM variants
    Decoding-time interpolation weight in Eqs. 2-5; the paper does not state how it was tuned. Results at a different lambda would shift the reported margins.
  • PHRASED selection weight mu and per-token bonus = 3.0 (streaming ST; from Section 5.1 text)
    Controls how strongly selected target phrases are boosted over the 1.0 list bonus. The 21% relative recall gain is measured at this value.
  • Phrase list composition (3K list with OntoNote5 distractors) = 3K phrases; not further specified
    Table 3 shows the PHRASED over PLB margin depends on list composition (11% relative on the oracle list vs 21% on the 3K list), so the headline numbers are attached to this hand-built list.
assumptions (5)
  • standard math Shallow fusion interpolation (Eq. 2) is a valid mechanism for injecting biasing scores into transducer decoding
    Inherited from ASR contextual biasing (Zhao et al. 2019, Wang et al. 2024), cited in Section 3.1. The paper adds no new justification for interpolation in ST decoding.
  • domain assumption The CTC output of CTC-GMM is accurate enough to match source-language phrases
    Section 3.3 uses the CTC output as the intermediate representation z for matching. A misrecognized rare phrase yields no match, no bonus, and a fallback to unassisted decoding, so this assumption caps the method's recall.
  • domain assumption The user-supplied {I:O} dictionary entries are correct and unambiguous
    Eqs. 4-5 boost the mapped target phrase whenever the source matches. Wrong or contextually inappropriate mappings would be injected into the output; the paper reports no filtering, confidence scoring, or analysis of dictionary errors.
  • ad hoc to paper The phrase recall metric, counting each target phrase once and excluding filler words, preserves method rankings
    Section 4.3 defines this evaluation convention. The authors note absolute recall values differ from duplication counting and assert comparisons are unaffected, but provide no sensitivity analysis.
  • ad hoc to paper The partial-match bonus cleanup in streaming decoding is exact under translation reordering
    Section 3.3 removes already-applied bonuses by multiplying the index of the longest partially matched phrase by the per-token bonus. The description is informal and no experiment isolates whether imperfect cleanup harms BLEU or recall.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PHRASED: Phrase Dictionary Biasing for Speech Translation." pith.science (2026). https://pith.science/paper/QZNFDL3A

@misc{pith2026250609175,
  author       = {Pith},
  title        = {Pith review of: PHRASED: Phrase Dictionary Biasing for Speech Translation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QZNFDL3A}},
  note         = {Machine review of arXiv:2506.09175}
}
read the original abstract

Phrases are essential to understand the core concepts in conversations. However, due to their rare occurrence in training data, correct translation of phrases is challenging in speech translation tasks. In this paper, we propose a phrase dictionary biasing method to leverage pairs of phrases mapping from the source language to the target language. We apply the phrase dictionary biasing method to two types of widely adopted models, a transducer-based streaming speech translation model and a multimodal large language model. Experimental results show that the phrase dictionary biasing method outperforms phrase list biasing by 21% relatively for the streaming speech translation model. In addition, phrase dictionary biasing enables multimodal large language models to use external phrase information, achieving 85% relative improvement in phrase recall.

Figures

Figures reproduced from arXiv: 2506.09175 by the authors.

Figure 1
Figure 1. Illustration of PHRASEDphrase_selection for CTC-GMM. At Step 1, we find the source￾language phrases that occur in the intermediate representation. At Step 2, we add bonuses to the ST model output for the selected target-language phrases [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Illustration of PHRASEDjoint_biasing for multimodal LLMs. At Step 1, we find the source-language phrases that occur in the intermediate representation z. At Step 2, we add the selected target-language phrases and related intermediate representation information to the prompt. 3.4 PHRASED for Multimodal LLM A multimodal LLM can support various speech functions in a single model. We first let the multimodal LLM convert… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

40 extracted references · 25 canonical work pages

  1. [1]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Abdelrahman Abouelenin, Atabak Ashfaq, Adam Atkinson, Hany Awadalla, Nguyen Bach, Jianmin Bao, Alon Benhaim, Martin Cai, Vishrav Chaudhary, Congcong Chen, et al. 2025. Phi-4-mini technical report: Compact yet powerful multimodal language models via mixture-of-loras. arXiv preprint arXiv:2503.01743

    arXiv 2025

  4. [4]

    Shanbo Cheng, Zhichao Huang, Tom Ko, Hang Li, Ningxin Peng, Lu Xu, and Qini Zhang. 2024. https://arxiv.org/abs/2407.21646 Towards achieving human parity on end-to-end simultaneous speech translation via llm agent . arXiv preprint arXiv:2407.21646

    arXiv 2024

  5. [5]

    Yexing Du, Ziyang Ma, Yifan Yang, Keqi Deng, Xie Chen, Bo Yang, Yang Xiang, Ming Liu, and Bing Qin. 2024. Cot-st: Enhancing llm-based speech translation with multimodal chain-of-thought. arXiv preprint arXiv:2409.19510

    arXiv 2024

  6. [6]

    Marco Gaido, Matteo Negri, and Marco Turchi. 2022. Who are we talking about? handling person names in speech translation. In Proceedings of the 19th International Conference on Spoken Language Translation (IWSLT 2022), pages 62--73

    work page 2022

  7. [7]

    Marco Gaido, Sara Papi, Matteo Negri, and Marco Turchi. 2023 a . Joint speech translation and named entity recognition. In Proc. Interspeech 2023, pages 47--51

    work page 2023

  8. [8]

    Marco Gaido, Yun Tang, Ilia Kulikov, Rongqing Huang, Hongyu Gong, and Hirofumi Inaguma. 2023 b . Named entity detection and injection for direct speech translation. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1--5. IEEE

    work page 2023

Show all 40 references
  1. [9]

    Sreyan Ghosh, Mohammad Sadegh Rasooli, Michael Levit, Peidong Wang, Jian Xue, Dinesh Manocha, and Jinyu Li. 2024. Failing forward: Improving generative error correction for asr with synthetic data and retrieval augmentation. arXiv preprint arXiv:2410.13198

    2024 arXiv

  2. [10]

    Hany Hassan, Anthony Aue, Chang Chen, Vishal Chowdhary, Jonathan Clark, Christian Federmann, Xuedong Huang, Marcin Junczys-Dowmunt, William Lewis, Mu Li, et al. 2018. Achieving human parity on automatic chinese to english news translation. arXiv preprint arXiv:1803.05567

    2018 arXiv

  3. [11]

    Ke Hu, Zhehuai Chen, Chao-Han Huck Yang, Piotr \.Z elasko, Oleksii Hrinchuk, Vitaly Lavrukhin, Jagadeesh Balam, and Boris Ginsburg. 2025. Chain-of-thought prompting for speech translation. In ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Proces...

    2025

  4. [12]

    Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, et al. 2024. Gpt-4o system card. arXiv preprint arXiv:2410.21276

    2024 arXiv

  5. [13]

    Jinyu Li. 2022. Recent advances in end-to-end automatic speech recognition. APSIPA Transactions on Signal and Information Processing, 11(1)

    2022

  6. [14]

    Xiaoqing Li, Jinghui Yan, Jiajun Zhang, and Chengqing Zong. 2019. Neural name translation improves neural machine translation. In Machine Translation: 14th China Workshop, CWMT 2018, Wuyishan, China, October 25-26, 2018, Proceedings 14, pages 93--100. Springer

    2019

  7. [15]

    Zhongwei Li, Xuancong Wang, AiTi Aw, Eng Siong Chng, and Haizhou Li. 2018. Named-entity tagging and domain adaptation for better customized translation. In Proceedings of the seventh named entities workshop, pages 41--46

    2018

  8. [16]

    Rao Ma, Mengjie Qian, Potsawee Manakul, Mark Gales, and Kate Knill. 2023. Can generative large language models perform asr error correction? arXiv preprint arXiv:2307.04172

    2023 arXiv

  9. [17]

    Evgeny Matusov, Stephan Kanthak, and Hermann Ney. 2005. On the integration of speech recognition and statistical machine translation. In Interspeech, pages 3177--3180

    2005

  10. [18]

    Maciej Modrzejewski, Miriam Exel, Bianka Buschbeck, Thanh-Le Ha, and Alex Waibel. 2020. Incorporating external annotation to improve named entity translation in nmt. In Proceedings of the 22nd annual conference of the european association for machine translation, pages 45--51

    2020

  11. [19]

    Hermann Ney. 1999. Speech translation: Coupling of recognition and translation. In Proceedings of ICASSP, pages 517--520

    1999

  12. [20]

    Jan Niehues and Eunah Cho. 2017. Exploiting linguistic resources for neural machine translation using multi-task learning. In Proceedings of the Second Conference on Machine Translation, pages 80--89

    2017

  13. [21]

    Matt Post, Shuoyang Ding, Marianna Martindale, and Winston Wu. 2019. An exploration of placeholding in neural machine translation. In Proceedings of Machine Translation Summit XVII: Research Track, pages 182--192

    2019

  14. [22]

    Matt Post, Gaurav Kumar, Adam Lopez, Damianos Karakos, Chris Callison-Burch, and Sanjeev Khudanpur. 2013. Improved speech-to-text translation with the fisher and callhome spanish-english speech translation corpus. In Proceedings of the 10th international workshop on spoken lan...

    2013

  15. [23]

    Rohit Prabhavalkar, Takaaki Hori, Tara N Sainath, Ralf Schl \"u ter, and Shinji Watanabe. 2023. End-to-end speech recognition: A survey. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 32:325--351

    2023

  16. [24]

    Sameer Pradhan, Alessandro Moschitti, Nianwen Xue, Hwee Tou Ng, Anders Bj \"o rkelund, Olga Uryupina, Yuchen Zhang, and Zhi Zhong. 2013. Towards robust linguistic analysis using ontonotes. In Proceedings of the Seventeenth Conference on Computational Natural Language Learning,...

    2013

  17. [25]

    Rico Sennrich and Barry Haddow. 2016. Linguistic input features improve neural machine translation. In Proceedings of the First Conference on Machine Translation: Volume 1, Research Papers, pages 83--91

    2016

  18. [26]

    Kai Song, Yue Zhang, Heng Yu, Weihua Luo, Kun Wang, and Min Zhang. 2019. Code-switching for enhancing nmt with pre-specified translation. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Techno...

    2019

  19. [27]

    Matthias Sperber and Matthias Paulik. 2020. Speech translation and the end-to-end promise: Taking stock of where we are. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7409--7421

    2020

  20. [28]

    Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, Katie Millican, et al. 2023. Gemini: a family of highly capable multimodal models. arXiv preprint arXiv:2312.11805

    2023 arXiv

  21. [29]

    Arata Ugawa, Akihiro Tamura, Takashi Ninomiya, Hiroya Takamura, and Manabu Okumura. 2018. Neural machine translation incorporating named entity. In Proceedings of the 27th International Conference on Computational Linguistics, pages 3240--3250

    2018

  22. [30]

    Laura Cross Vila, Carlos Escolano, Jos \'e AR Fonollosa, and Marta R Costa-Jussa. 2018. End-to-end speech translation with the transformer. In Proceedings of Interspeech, pages 60--63

    2018

  23. [31]

    Peidong Wang, Eric Sun, Jian Xue, Yu Wu, Long Zhou, Yashesh Gaur, Shujie Liu, and Jinyu Li. 2023. Lamassu: Streaming language-agnostic multilingual speech recognition and translation using neural transducers. In INTERSPEECH 2023, pages 57--61

    2023

  24. [32]

    Weiran Wang, Zelin Wu, Diamantino Caseiro, Tsendsuren Munkhdalai, Khe Chai Sim, Pat Rondon, Golan Pundak, Gan Song, Rohit Prabhavalkar, Zhong Meng, et al. 2024. Contextual biasing with the knuth-morris-pratt matching algorithm. In Proc. Interspeech 2024, pages 282--286

    2024

  25. [33]

    Yuguang Wang, Shanbo Cheng, Liyang Jiang, Jiajun Yang, Wei Chen, Muze Li, Lin Shi, Yanfeng Wang, and Hongtao Yang. 2017. Sogou neural machine translation systems for wmt17. In Proceedings of the Second Conference on Machine Translation, pages 410--415

    2017

  26. [34]

    Shufang Xie, Yingce Xia, Lijun Wu, Yiqing Huang, Yang Fan, and Tao Qin. 2022. End-to-end entity-aware neural machine translation. Machine Learning, 111(3):1181--1203

    2022

  27. [35]

    Jian Xue, Peidong Wang, Jinyu Li, Matt Post, and Yashesh Gaur. 2022. Large-scale streaming end-to-end speech translation with neural transducers. arXiv preprint arXiv:2204.05352

    2022 arXiv

  28. [36]

    Jian Xue, Peidong Wang, Jinyu Li, and Eric Sun. 2023. A weakly-supervised streaming multilingual speech model with truly zero-shot capability. In 2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), pages 1--7. IEEE

    2023

  29. [37]

    Chao-Han Huck Yang, Yile Gu, Yi-Chieh Liu, Shalini Ghosh, Ivan Bulyko, and Andreas Stolcke. 2023. Generative speech recognition error correction with large language models and task-activating prompting. In 2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU...

    2023

  30. [38]

    Ding Zhao, Tara N Sainath, David Rybach, Pat Rondon, Deepti Bhatia, Bo Li, and Ruoming Pang. 2019. Shallow-fusion end-to-end contextual biasing. In Interspeech, pages 1418--1422

    2019

  31. [39]

    Penghao Zhao, Hailin Zhang, Qinhan Yu, Zhengren Wang, Yunteng Geng, Fangcheng Fu, Ling Yang, Wentao Zhang, Jie Jiang, and Bin Cui. 2024 a . Retrieval-augmented generation for ai-generated content: A survey. arXiv preprint arXiv:2402.19473

    2024 arXiv

  32. [40]

    Rui Zhao, Jinyu Li, Ruchao Fan, and Matt Post. 2024 b . Ctc-gmm: Ctc guided modality matching for fast and accurate streaming speech translation. In 2024 IEEE Spoken Language Technology Workshop (SLT), pages 1068--1075. IEEE

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.