Pith. sign in

REVIEW 3 major objections 3 minor 34 references

How to Recognize New Words: A Comparison Between Context Biasing Methods and Speech LLMs

T0 review · 3 major / 3 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read When both strategies can take the same word list, dedicated context biasing is the more reliable way to recognize new and rare words, especially on non-read speech.

desk verdict A genuinely useful empirical comparison, but the paper asserts prompt-order sensitivity without showing the data, and its alphabetical ordering could be loading the dice against speech LLMs. read the letter →

arxiv 2608.05759 v1 pith:LSTVOYGM submitted 2026-08-06 cs.CL

Christian Huber , Alexander Waibel This is my paper · ORCID
classification cs.CL
keywords contextbiasingspeechLLMsrarewordrecognitionautomaticerrorratepromptsensitivitydistractorrobustness
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

The paper sets out to decide which of two ways to help an ASR system recognize new and rare words—names, acronyms, domain terms—works better in practice: dedicated context biasing, where a word list is fed into a modified Whisper model, or speech large language models prompted with the same list. It claims that the dedicated methods cut the error rate on the biased words by up to 88% relative while leaving other words largely unaffected, and that they tolerate noisy, long word lists far better than speech LLMs do. The speech LLMs, it argues, are strong on read speech but generalize less well to conversations and meetings, and their accuracy depends on how many distractor words share the prompt and in what order the words appear. This matters because rare words usually carry the informative content of an utterance, so knowing which technique can actually write them correctly determines whether a transcript is usable.

What carries the argument

The load-bearing object is the context biasing list $Z=(Z_1,\ldots,Z_L)$ supplied at inference. Method A adds a context-attention layer after each decoder cross-attention layer: for each decoded token it computes similarity between the decoder state and summary vectors of the list entries, and with an $\mathrm{argmax}$ selects the single most relevant entry to attend to; if none is relevant, it uses a learned dummy vector. Method B extends the decoder's vocabulary with dynamic tokens $v_l$ for list entries, embedding them from the same summary vectors and scoring them with a scaled dot product between the decoder output and the context summaries. The speech LLMs instead receive the words as text tokens in the prompt, which is why prompt order and list length affect them.

What would settle it

Re-run the Earnings-21 and Yodas experiments with speech-LLM prompts ordered by placing the target word first instead of alphabetically; if a natural ordering brings their biased WER at 250 distractors down to or below method B's, the paper's robustness conclusion is falsified.

Watch

Extended reading notes

Core claim

On its own terms, the paper's discovery is a trade-off map rather than a single winner. Against Whisper large-v2, the two context biasing methods improve biased WER by 37–70% relative for method A and 48–88% relative for method B when the correct word is in the list, while unbiased WER moves by at most a few percent. Method B is the best non-LLM overall, but loses ground as distractors are added because it wrongly inserts dynamic vocabulary tokens. The three speech LLMs are competitive or better on read LibriSpeech—Qwen3-Omni reaches 1.1% biased WER on test-clean—but fall behind on Earnings-21 and Yodas, and their biased WER degrades by 39% to 570% relative when 250 distractors are added. Filtering the word list with an auxiliary speech LLM recovers much of the loss but does not close the gap on non-read speech.

Load-bearing premise

The comparison depends on treating alphabetically sorted prompts as the representative way to prompt speech LLMs; if some natural ordering systematically improves their rare-word accuracy, the conclusion that dedicated context biasing is more robust could change.

Editorial extensions

If this is right

  • With a clean, short word list, dedicated context biasing—especially method B—attains lower biased WER on non-read speech than any of the three speech LLMs tested.
  • Adding up to 250 unrelated distractor words costs context biasing at most a few percent relative, while it can multiply speech-LLM biased WER several-fold; applications with long lists should expect the gap to widen.
  • Speech LLMs remain attractive for read-style audio where their unbiased baselines are already strong; Qwen3-Omni's 1.1% biased WER on LibriSpeech test-clean is the best single number in the table.
  • Filtering the list with an auxiliary model restores most of the speech-LLM loss at high recall, but leaves them behind method B on Earnings-21 and Yodas, so filtering is a partial mitigation, not a fix.
  • Since speech LLMs are word-order sensitive, any deployment that feeds a list must control prompt order; alphabetical sorting is the paper's controlled choice, not an instruction for practice.

Reading between the lines

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

  • A practical extension the paper does not test: use the positional sensitivity itself as a lever—place the most important or hardest words first in the prompt, or order by acoustic confusability—and measure whether speech LLMs close the gap on non-read speech.
  • Because the paper intentionally isolates word-list prompting, its verdict on speech LLM flexibility is limited to that interface; in settings where context can be a free-form description, the comparison may look different, and the paper notes this capability but does not measure it.
  • A testable prediction follows from the distractor results: speech-LLM biased WER should degrade monotonically with prompt length and with later word position, while context biasing should stay roughly flat; a practitioner could build a cheap pre-filter that orders prompts by estimated target probability and verify the predicted gap.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 3 minor

Summary. The manuscript compares two context biasing extensions of Whisper (methods A and B) with three speech LLMs (Qwen3-ASR, Qwen3-Omni, VibeVoice-ASR) on rare-word recognition. It reports biased, unbiased, and overall word error rates (BWER, UWER, WER) on Earnings-21, LibriSpeech test-clean/test-other, and Yodas, with 0, 10, 100, and 250 distractor words. It also evaluates a filtering stage that uses Qwen3-Omni to prune the bias list. The headline claims are that context biasing cuts biased WER by up to 88% relative, leaves other words largely unaffected, and remains robust to distractors, while speech LLMs are strong on read speech but weaker on non-read speech and sensitive to distractor count and prompt word order.

Significance. If the results hold, the paper offers practical guidance for selecting between dedicated context biasing and speech-LLM prompting. The strengths are the controlled distractor-count sweep, the use of three public test sets covering read and non-read speech, the clear metric definitions, and the filtering study with recall/precision reporting. The paper is honest about limitations, including the computational cost of filtering and the fact that free-form prompting is not tested. The main concern is that the speech-LLM comparison rests on a single prompt-ordering choice that is not experimentally justified, and no statistical uncertainty is reported for the key comparisons.

major comments (3)
  1. [Section IV-A and Table I] The speech-LLM columns are all produced with alphabetical sorting of the prompt words, while the paper asserts without supporting experiment that speech LLMs are order-sensitive and perform better when a word is at the beginning of the prompt. Since the central conclusion that context biasing is more robust than speech LLMs depends on these columns, the missing order experiment is load-bearing. Please add a controlled comparison of at least two alternative orderings (e.g., random order and relevant-word-first) on Earnings-21 and Yodas at N=0 and N=250, and report the resulting BWER values.
  2. [Section V and Table I] No confidence intervals, significance tests, or repeated distractor draws are provided. Claims such as “method B is better than all speech LLMs” at N=10 and “UWER essentially unchanged” are based on single test-set numbers, and the distractors are described as “chosen randomly” without a seed or repeated sampling. Add bootstrap confidence intervals over utterances and/or multiple distractor draws so the reader can assess whether the reported gaps are significant.
  3. [Section V, Table I, earnings and Yodas rows] The text states that for models C, D, and E “the BWER performance without relevant context does not change significantly when adding distractors,” but the table shows relative changes of roughly 5–15% (for example, Earnings-21, model D: 30.81 at N=0 vs. 26.29 at N=250). Without a statistical test, “significantly” is unsupported; rephrase as a magnitude statement or add significance tests.
minor comments (3)
  1. [Throughout] There are typographical issues: “VibeV oice-ASR” has an unintended space, “context biassing” appears in Section V, and Table I mixes commas and periods as decimal separators.
  2. [Section IV-A] The sentence “We found that they are sensitive with respect to the order in which the words from the context biasing list are placed in the prompt” reports a finding without any accompanying data; either provide the supporting experiment or rephrase it as a hypothesis.
  3. [Section I and Section V] The introduction lists “quantifying sensitivity to distractors and prompt word order” as a contribution, but no prompt-order result appears in Section V; align the stated contributions with the experiments actually reported.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the comparison rests on external test sets and direct evaluation, not on fitted or self-cited premises.

full rationale

This is a measurement/comparison paper. The headline claims (BWER reductions up to 88%, speech-LLM sensitivity to distractor count and prompt order) are direct empirical observations on Earnings-21, LibriSpeech, and Yodas, with no parameter fitted to the reported outcome. Methods A and B are trained on Common Voice and evaluated on held-out test sets, and the speech LLMs are used as provided. The only appearance of the authors' own prior work is [12], which supplies the training scheme and one architecture for method A; this is methodological reuse rather than a load-bearing theoretical premise. The alphabetical prompt sorting is a stated protocol choice made after observing order sensitivity; whether it is the fairest possible prompt is a fairness or robustness concern, not a circularity, because the central conclusion is not defined in terms of that sorting. No self-citation chain, uniqueness claim, fitted parameter, or by-construction equivalence is used to force the reported comparison. Therefore no circular step is exhibited.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

No new entities, forces, or conserved quantities are introduced; the paper reuses existing models and methods. The central claims rest on the representativeness of the chosen models and test sets and on the fairness of the alphabetical prompt ordering, plus standard ASR and attention math from prior work.

free parameters (5)
  • Rare-word frequency cutoff = top 10%
    LibriSpeech and Yodas rare words are defined as words outside the 10% most common words; this hand-chosen cutoff determines which words count as rare and therefore shapes the test sets and resulting WERs.
  • Minimum utterance frequency for rare words = 2 (LibriSpeech), 4 (Yodas)
    Words must occur in at least two utterances for LibriSpeech or four occurrences within one YouTube video for Yodas; these thresholds filter noise but are arbitrary choices affecting the test sets.
  • Distractor count = 0, 10, 100, 250
    The number of distractors added to the context list is a hand-chosen evaluation variable; the distractor sensitivity conclusion depends on this range.
  • Loss balancing hyperparameter lambda = not reported
    Method A's training loss uses lambda to weight the context-selection cross-entropy term; the value is not given, making replication difficult.
  • Context list length during training = 200 entries
    The context biasing list is filled up to 200 entries with distractors during training; this choice may affect distractor robustness at evaluation N=250.
assumptions (4)
  • domain assumption Alphabetical ordering of prompt words is a fair and representative prompting strategy for speech LLMs.
    The authors find speech LLMs are order-sensitive and sort words alphabetically to ensure fairness; if a different ordering systemically improves speech LLM performance, the comparison to context biasing changes. Stated in Section IV-A.
  • domain assumption The rare-word definitions (outside top 10%, occurrence thresholds) yield a valid operationalization of 'new and rare words' and the recognition difficulty they cause.
    The hand-crafted filters in Section IV-B determine which words count as rare; the results are conditional on these choices.
  • domain assumption The evaluated models are representative of their respective classes: two Whisper-based biasing methods represent context biasing, three speech LLMs represent speech LLMs.
    Conclusions about 'context biasing methods' and 'speech LLMs' in general are drawn from five specific models; scale and training data differ across classes, acknowledged in Section V-A.
  • standard math Standard sequence-to-sequence and attention equations hold as in the cited literature.
    The paper builds on standard transformer ASR and attention mechanisms (Equations 1-5) without proving them.

how reviews work

0 comments
Cite this review

Pith. "Pith review of How to Recognize New Words: A Comparison Between Context Biasing Methods and Speech LLMs." pith.science (2026). https://pith.science/paper/LSTVOYGM

@misc{pith2026260805759,
  author       = {Pith},
  title        = {Pith review of: How to Recognize New Words: A Comparison Between Context Biasing Methods and Speech LLMs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LSTVOYGM}},
  note         = {Machine review of arXiv:2608.05759}
}
read the original abstract

Recognizing new and rare words - named entities, acronyms, domain specific special words, and other items scarce in training data - remains a key challenge for automatic speech recognition (ASR). We compare two strategies for this: context biasing methods, where an ASR model is extended such that during inference a word list can be supplied, and speech large language models (LLMs) prompted with context directly. We evaluate two context biasing methods based on Whisper against three speech LLMs across read and non-read speech, reporting biased, unbiased, and overall word error rate (WER). The context biasing methods cut biased WER by up to 88% relative while leaving other words largely unaffected. Speech LLMs excel on read speech but generalize less well to non-read speech, and prove sensitive to distractor count and prompt word order. We characterize the resulting trade-offs to guide method selection.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 19 canonical work pages

  1. [1]

    Attention is all you need,

    A. Vaswani, “Attention is all you need,”Advances in Neural Information Processing Systems, 2017

    2017

  2. [2]

    Very deep self-attention networks for end-to-end speech recognition,

    N.-Q. Pham, T.-S. Nguyen, J. Niehues, M. M ¨uller, S. St ¨uker, and A. Waibel, “Very deep self-attention networks for end-to-end speech recognition,”arXiv preprint arXiv:1904.13377, 2019

    arXiv 1904

  3. [3]

    Robust speech recognition via large-scale weak supervi- sion,

    A. Radford, J. W. Kim, T. Xu, G. Brockman, C. McLeavey, and I. Sutskever, “Robust speech recognition via large-scale weak supervi- sion,” inInternational conference on machine learning. PMLR, 2023

    2023

  4. [4]

    Cold fusion: Training seq2seq models together with language models,

    A. Sriram, H. Jun, S. Satheesh, and A. Coates, “Cold fusion: Training seq2seq models together with language models,”Interspeech 2018, 2018

    work page 2018

  5. [5]

    Contextual speech recognition in end-to-end neural network systems using beam search

    I. Williams, A. Kannan, P. S. Aleksic, D. Rybach, and T. N. Sainath, “Contextual speech recognition in end-to-end neural network systems using beam search.” inInterspeech, 2018, pp. 2227–2231

    work page 2018

  6. [6]

    An analysis of incorporating an external language model into a sequence-to-sequence model,

    A. Kannan, Y . Wu, P. Nguyen, T. N. Sainath, Z. Chen, and R. Prab- havalkar, “An analysis of incorporating an external language model into a sequence-to-sequence model,” in2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2018, pp. 1–5828

    work page 2018

  7. [7]

    Class lm and word mapping for contextual biasing in end-to-end asr,

    R. Huang, O. Abdel-hamid, X. Li, and G. Evermann, “Class lm and word mapping for contextual biasing in end-to-end asr,”Interspeech 2020, 2020

    work page 2020

  8. [8]

    A study of biasing technical terms in medical speech recognition using weighted finite-state transducer,

    A. Kojima, “A study of biasing technical terms in medical speech recognition using weighted finite-state transducer,”Acoustical Science and Technology, vol. 43, no. 1, pp. 66–68, 2022

    work page 2022

Show all 34 references
  1. [9]

    Deep context: end-to-end contextual speech recognition,

    G. Pundak, T. N. Sainath, R. Prabhavalkar, A. Kannan, and D. Zhao, “Deep context: end-to-end contextual speech recognition,” in2018 IEEE spoken language technology workshop (SLT). IEEE, 2018, pp. 418–425

    2018

  2. [10]

    Phoebe: Pronunciation-aware contextualization for end-to-end speech recogni- tion,

    A. Bruguier, R. Prabhavalkar, G. Pundak, and T. N. Sainath, “Phoebe: Pronunciation-aware contextualization for end-to-end speech recogni- tion,” inIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2019

    2019

  3. [11]

    Contextual rnn-t for open domain asr,

    M. Jain, G. Keren, J. Mahadeokar, G. Zweig, F. Metze, and Y . Saraf, “Contextual rnn-t for open domain asr,” 2020

    2020

  4. [12]

    Instant one-shot word- learning for context-specific neural sequence-to-sequence speech recog- nition,

    C. Huber, J. Hussain, S. St ¨uker, and A. Waibel, “Instant one-shot word- learning for context-specific neural sequence-to-sequence speech recog- nition,” in2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU). IEEE, 2021, pp. 1–7

    2021

  5. [13]

    Contextualized streaming end-to-end speech recognition with trie-based deep biasing and shallow fusion,

    D. Le, M. Jain, G. Keren, S. Kim, Y . Shi, J. Mahadeokar, J. Chan, Y . Shangguan, C. Fuegen, O. Kalinliet al., “Contextualized streaming end-to-end speech recognition with trie-based deep biasing and shallow fusion,”arXiv preprint arXiv:2104.02194, 2021

    2021 arXiv

  6. [14]

    Im- proving end-to-end contextual speech recognition with fine-grained con- textual knowledge selection,

    M. Han, L. Dong, Z. Liang, M. Cai, S. Zhou, Z. Ma, and B. Xu, “Im- proving end-to-end contextual speech recognition with fine-grained con- textual knowledge selection,” inICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, ...

    2022

  7. [15]

    Personalization of ctc speech recognition models,

    S. Dingliwal, M. Sunkara, S. Ronanki, J. Farris, K. Kirchhoff, and S. Bodapati, “Personalization of ctc speech recognition models,” in2022 IEEE Spoken Language Technology Workshop (SLT). IEEE, 2023, pp. 302–309

    2023

  8. [16]

    Contextualized end-to-end speech recognition with contextual phrase prediction network,

    K. Huang, A. Zhang, Z. Yang, P. Guo, B. Mu, T. Xu, and L. Xie, “Contextualized end-to-end speech recognition with contextual phrase prediction network,” inProc. Interspeech 2023, 2023, pp. 4933–4937

    2023

  9. [17]

    Promptasr for contextualized asr with controllable style,

    X. Yang, W. Kang, Z. Yao, Y . Yang, L. Guo, F. Kuang, L. Lin, and D. Povey, “Promptasr for contextualized asr with controllable style,” in ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024, pp. 10 536–10 540

    2024

  10. [18]

    Contex- tualized automatic speech recognition with attention-based bias phrase boosted beam search,

    Y . Sudo, M. Shakeel, Y . Fukumoto, Y . Peng, and S. Watanabe, “Contex- tualized automatic speech recognition with attention-based bias phrase boosted beam search,” inICASSP 2024-2024 IEEE International Con- ference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 20...

    2024

  11. [19]

    Lcb-net: Long-context bias- ing for audio-visual speech recognition,

    F. Yu, H. Wang, X. Shi, and S. Zhang, “Lcb-net: Long-context bias- ing for audio-visual speech recognition,” inICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024

    2024

  12. [20]

    Contextual asr with retrieval augmented large language model,

    C. Xiao, Z. Hou, D. Garcia-Romero, and K. J. Han, “Contextual asr with retrieval augmented large language model,” inICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2025, pp. 1–5

    2025

  13. [21]

    Owsm-biasing: Contextualizing open whisper-style speech models for automatic speech recognition with dynamic vocabulary,

    Y . Sudo, Y . Fujita, A. Kojima, T. Mizumoto, and L. Liu, “Owsm-biasing: Contextualizing open whisper-style speech models for automatic speech recognition with dynamic vocabulary,”arXiv preprint arXiv:2506.09448, 2025

    2025 arXiv

  14. [22]

    Qwen-audio: Advancing universal audio understanding via unified large-scale audio-language models,

    Y . Chu, J. Xu, X. Zhou, Q. Yang, S. Zhang, Z. Yan, C. Zhou, and J. Zhou, “Qwen-audio: Advancing universal audio understanding via unified large-scale audio-language models,”arXiv preprint arXiv:2311.07919, 2023

    2023 arXiv

  15. [23]

    Salmonn: Towards generic hearing abilities for large language models,

    C. Tang, W. Yu, G. Sun, X. Chen, T. Tan, W. Li, L. Lu, Z. Ma, and C. Zhang, “Salmonn: Towards generic hearing abilities for large language models,” inInternational Conference on Learning Represen- tations, vol. 2024, 2024, pp. 16 607–16 629

    2024

  16. [24]

    Wavllm: Towards robust and adaptive speech large language model,

    S. Hu, L. Zhou, S. Liu, S. Chen, L. Meng, H. Hao, J. Pan, X. Liu, J. Li, S. Sivasankaranet al., “Wavllm: Towards robust and adaptive speech large language model,” inFindings of the Association for Computational Linguistics: EMNLP 2024, 2024, pp. 4552–4572

    2024

  17. [25]

    End-to-end speech recognition contextualization with large language models,

    E. Lakomkin, C. Wu, Y . Fathullah, O. Kalinli, M. L. Seltzer, and C. Fuegen, “End-to-end speech recognition contextualization with large language models,” inICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024, pp. 12 4...

    2024

  18. [26]

    Contextual biasing speech recognition in speech-enhanced large language model

    X. Gong, A. Lv, Z. Wang, and Y . Qian, “Contextual biasing speech recognition in speech-enhanced large language model.” inINTER- SPEECH, 2024

    2024

  19. [27]

    Qwen3-asr technical report,

    X. Shi, X. Wang, Z. Guo, Y . Wang, P. Zhang, X. Zhang, Z. Guo, H. Hao, Y . Xi, B. Yang, J. Xu, J. Zhou, and J. Lin, “Qwen3-asr technical report,” arXiv preprint arXiv:2601.21337, 2026

    2026 arXiv

  20. [28]

    Qwen3-omni technical report,

    J. Xu, Z. Guo, H. Hu, Y . Chu, X. Wang, J. He, Y . Wang, X. Shi, T. He, X. Zhuet al., “Qwen3-omni technical report,”arXiv preprint arXiv:2509.17765, 2025

    2025 arXiv

  21. [29]

    Vibevoice-asr technical report,

    Z. Peng, J. Yu, Y . Chang, Z. Wang, L. Dong, Y . Hao, Y . Tu, C. Yang, W. Wang, S. Xuet al., “Vibevoice-asr technical report,”arXiv preprint arXiv:2601.18184, 2026

    2026

  22. [30]

    Multilingual translation with extensible multilingual pretraining and finetuning,

    Y . Tang, C. Tran, X. Li, P.-J. Chen, N. Goyal, V . Chaudhary, J. Gu, and A. Fan, “Multilingual translation with extensible multilingual pretraining and finetuning,”arXiv preprint arXiv:2008.00401, 2020

    2008 arXiv

  23. [31]

    Com- mon voice: A massively-multilingual speech corpus,

    R. Ardila, M. Branson, K. Davis, M. Henretty, M. Kohler, J. Meyer, R. Morais, L. Saunders, F. M. Tyers, and G. Weber, “Com- mon voice: A massively-multilingual speech corpus,”arXiv preprint arXiv:1912.06670, 2019

    1912 arXiv

  24. [32]

    Earnings-21: A practical benchmark for asr in the wild,

    M. Del Rio, N. Delworth, R. Westerman, M. Huang, N. Bhandari, J. Palakapilly, Q. McNamara, J. Dong, P. Zelasko, and M. Jett ´e, “Earnings-21: A practical benchmark for asr in the wild,”arXiv preprint arXiv:2104.11348, 2021

    2021 arXiv

  25. [33]

    Librispeech: an asr corpus based on public domain audio books,

    V . Panayotov, G. Chen, D. Povey, and S. Khudanpur, “Librispeech: an asr corpus based on public domain audio books,” in2015 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, 2015, pp. 5206–5210

    2015

  26. [34]

    Yodas: Youtube-oriented dataset for audio and speech,

    X. Li, S. Takamichi, T. Saeki, W. Chen, S. Shiota, and S. Watanabe, “Yodas: Youtube-oriented dataset for audio and speech,” in2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU). IEEE, 2023, pp. 1–8

    2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.