Enhancing Whisper's Accuracy and Speed for Indian Languages through Prompt-Tuning and Tokenization

Kumud Tripathi; Pankaj Wasnik; Raj Gothi

REVIEW 5 major objections 7 minor 21 references

Enhancing Whisper's Accuracy and Speed for Indian Languages through Prompt-Tuning and Tokenization

T0 review · 5 major / 7 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Adding language-family prompts and per-language tokenizer tokens improves Whisper's accuracy and cuts inference time on eight Indian languages.

desk verdict The accuracy claims are confounded by unequal training epochs; the speed idea is plausible but the paper needs major revision before the results can be trusted. read the letter →

arxiv 2412.19785 v1 pith:7IMPVU4M submitted 2024-12-27 cs.CL eess.AS

Kumud Tripathi , Raj Gothi , Pankaj Wasnik This is my paper · ORCID

classification cs.CLeess.AS

keywords automaticspeechrecognitionWhisperprompttuningtokenizerIndianlanguageslow-resourceinferencespeedworderrorrate

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

This paper tries to establish that Whisper can be made both more accurate and much faster on eight Indian languages by two cheap modifications that touch only its token interface. The first is prompt-tuning with language-family information: the decoder is given a token such as for Indo-Aryan languages or for Dravidian languages, so linguistically related languages can share phonetic and structural knowledge during fine-tuning. The second is a tokenizer extension that adds 250 language-specific byte-pair-encoding tokens per language, shrinking the number of tokens each transcription requires and therefore the number of autoregressive decoding steps. On the paper's Kathbath experiments, prompt-tuning yields the best word error rate, the tokenizer yields the fastest inference, and the two together give a tunable balance between accuracy and speed. This would matter because it offers a low-cost path to making a general-purpose multilingual ASR model serve low-resource languages without retraining the architecture.

What carries the argument

The load-bearing mechanism is the decoder token interface of Whisper. At inference, Whisper conditions on a fixed token prefix (<SOT>, language ID, task, no-timestamps); the paper inserts an extra learned token <SOTP> followed by a family prompt (<indo> or <dra>) before this prefix, so the decoder can condition on family-level phonetic regularities. For speed, the paper extends the byte-pair-encoding tokenizer that Whisper reuses, adding 250 new BPE tokens per language and widening the final token-head projection with random rows for those tokens while keeping the original weights. The tokenizer change compresses each output sentence into fewer tokens, and because decoding is autoregressive, fewer tokens means fewer decoder steps; the prompt change re-routes the decoder's attention through shared family structure, and that shared structure is what carries the accuracy gain.

What would settle it

Run the same prompt-tuning and tokenizer modifications on a held-out Indian-language corpus that was not used to pick the token count, or on Indo-Aryan and Dravidian languages outside the tested eight. The claim would be falsified if, on those languages, removing the family prompt does not raise WER or adding 250 tokens per language does not reduce generated-token counts and decoding time.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that Whisper's multilingual decoder can be steered toward Indian languages by two complementary interventions that touch only its token interface. Inserting a learned language-family prompt (<indo> or <dra>) before the start-of-transcription token during fine-tuning encodes the linguistic similarity shared by Hindi, Gujarati, Marathi, and Bengali on one side and Tamil, Telugu, Kannada, and Malayalam on the other, cutting average WER from the multilingual fine-tuned baseline of 20.15% to 18.53% in the best prompt-only configuration on Whisper Medium. Replacing the stock tokenizer with one augmented by per-language BPE tokens (250 per language) shrinks the number of generated tokens—27 to 19 for a Hindi sentence and 79 to 31 for Malayalam—and with it the wall-clock inference time, from 45.15 to 23.75 minutes on average for the fine-tuned Medium model. Using both modifications together retains most of the WER gain and most of the speed gain, which the paper presents as an accuracy-latency trade-off the user can adjust.

Load-bearing premise

The load-bearing premise is that 250 added tokens per language—the number that scored best on the same Kathbath test set whose results are reported—and the two coarse family prompts are generalizable choices for other Indian languages and datasets; the paper does not validate them on held-out languages or with significance tests.

Editorial extensions

If this is right

With prompt-tuning alone, Whisper Medium's average WER drops from 20.15% (multilingual fine-tuning) to 18.53%, and the paper reports lower WER for every one of the eight languages.
With the 250-token-per-language tokenizer alone, average inference time falls from 45.15 to 23.75 minutes on the Kathbath evaluation set while average WER stays close to the fine-tuned baseline at 19.67%.
Applying both techniques gives an intermediate operating point—20.05% WER and 39.35 minutes—so users can choose between the most accurate model, the fastest model, or a balance.
The tokenizer shortens a Hindi sentence from 27 to 19 generated tokens and a Malayalam sentence from 79 to 31 tokens, and the paper reports similar speed and WER benefits when the tokenizer is added to Whisper Small and to two other fast-inference Whisper baselines.
Prompt-tuning also helps smaller and larger models: average WER improves from 24.54% to 23.81% on Whisper Small and from 18.16% to 17.15% on Whisper Large.

Reading between the lines

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

A natural extension the paper does not run is to test the same two modifications on languages from other families, such as Tibeto-Burman or Austroasiatic languages of India; if the family-prompt gain is real, it should transfer to any family with enough shared phonetics, not just Indo-Aryan and Dravidian.
Because the decoder's cost scales with the number of tokens, the tokenizer's speed advantage should grow with utterance length and in streaming or real-time settings; the reported batch timings may understate that benefit.
The token count was selected on the same Kathbath test set used for final reporting, so the claimed 250-token optimum is likely optimistic; a held-out development set would be needed to know the true out-of-sample optimum.
The two mechanisms are not shown to be additive in the paper—the combined model is slightly worse than the prompt-only model in WER and worse than the tokenizer-only model in speed—so the balance claim is best read as an engineering trade-off rather than evidence of complementary gains.

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.

Desk Editor's Note

read the letter

The one thing to know before reading this paper is that its central WER comparisons are not controlled. Section V says all proposed prompt- and tokenizer-based models are fine-tuned for five epochs, while all remaining models get three. That means the W-M FT baseline in Table I and the "w/o Prompt" columns in Table V were trained for fewer steps than the proposed models. The reported accuracy gains from prompt-tuning and the tokenizer could just be extra gradient steps. That is a load-bearing flaw, not a minor omission.

The speed side of the story is more plausible. Table III shows that adding BPE tokens cuts generated token counts substantially for Hindi and Malayalam, which should directly reduce autoregressive decoding time. But Table IV is again confounded: the "w/ Tok" rows for WhisperX and Faster Whisper show WER dropping from roughly 87% to 19%, which makes sense only if those models were fine-tuned, while the "w/o Tok" baselines are presumably off-the-shelf. That is not a controlled comparison.

To give credit: the combination of language-family prompts (/) with BPE extension for eight Indian languages on Kathbath is a reasonable application of known techniques. The paper is clearly written and the token-count intuition is useful. But the methods themselves are incremental: prompt tuning for Whisper is in [7] and [15], and BPE extension is in [8] and [17]. The novelty is mostly in the specific combination and the evaluation.

Other soft spots: y=250 is chosen on the same test set used to report final numbers, so the token-count sweep in Table II is in-sample selection. The abstract's claim of accuracy gains across all model sizes is contradicted by the Small results in Table V, where Marathi, Bengali, and Malayalam degrade with the prompt. There are no error bars or significance tests anywhere, and no code or detailed tokenizer training procedure is released.

This paper deserves a serious referee, but only with the expectation of major revision. The fixes are straightforward: train all baselines for five epochs, validate y on held-out data, run a properly fine-tuned WhisperX/Faster Whisper baseline, and report confidence intervals. If the authors do that, the result could be a modest but useful engineering recipe for low-resource Indian-language ASR. As is, I would not cite it for the accuracy numbers, and I would keep the speed claim tentative. Send it to review with those conditions, or desk reject if the authors are unwilling to rerun the experiments.

Referee Report

5 major / 7 minor

Summary. The manuscript proposes two modifications to OpenAI's Whisper for automatic speech recognition of eight Indian languages (four Indo-Aryan and four Dravidian) using the Kathbath dataset. The first modification is prompt-tuning with language-family tokens (<indo> and <dra>) prepended to the decoder input. The second is an extended BPE tokenizer that adds per-language tokens to Whisper's token head. The authors report experiments on Whisper Small, Medium, and Large, claiming that prompt-tuning improves WER, the tokenizer reduces inference time by cutting the number of generated tokens, and the combined model balances accuracy and speed.

Significance. If the empirical claims were supported, the paper would offer a practical, low-cost recipe for adapting a strong multilingual ASR model to related low-resource languages: family-level prompts require only two new tokens, and a small tokenizer extension reduces autoregressive decoding steps. The idea of exploiting language family structure through prompts is well motivated and potentially useful. However, the evaluation as presented is compromised by an unequal training budget between baselines and proposed models, test-set-based selection of the token count, and a complete lack of statistical significance testing; these issues presently prevent the paper from establishing its central accuracy claims. The inference-speed part of the claim is more plausible because it is tied to the mechanical reduction in generated tokens, but even that comparison is not cleanly reported.

major comments (5)

[Section V, Experimental Setup] The sentence 'All the proposed prompting-based and tokenizer-based Whisper models are fine-tuned for five epochs and three epochs for all the remaining models' means that the W-M FT, W-S FT, W-L FT baselines and the w/o Tok variants in Table IV are trained for three epochs, while every proposed variant is trained for five epochs. As a result, the WER improvements in Tables I, IV, and V are confounded with additional gradient steps; a three-epoch baseline is not an adequate control. This confound directly affects the paper's central claims that prompt-tuning improves accuracy and that the combined model achieves an optimal balance between WER and inference speed. The authors must retrain all models with a matched number of epochs (or matched number of optimizer steps) and re-report all accuracy tables.
[Table II and Section V] The number of added tokens per language, y=250, is selected because it yields the best average WER on the same Kathbath test set used to report final results. This is a test-set selection bias: the reported 19.67% average for y=250 is the minimum over the four values {1000, 500, 250, 125} evaluated on the test set, so it is an optimistic estimate of generalization for that hyperparameter. Please select y on a held-out validation set or via nested cross-validation, and describe the selection procedure explicitly.
[Table IV] The comparison to WhisperX and Faster Whisper is not controlled for fine-tuning. The w/o Tok columns of Table IV show very high WER (e.g., 109.54% for Gujarati with WhisperX and 111.54% with Faster Whisper), which is consistent with no fine-tuning on the target languages, while the w/ Tok columns show much lower WER (14.99% and 26.22% respectively) after presumably fine-tuning with the extended tokenizer. Any improvement due to training on Kathbath is thus being attributed to the tokenizer. The authors must state explicitly whether the w/ Tok WhisperX/Faster Whisper models were fine-tuned, with what data, for how many epochs, and either match the training protocol across all systems or restrict the tokenizer comparison to models trained under identical conditions.
[Abstract and Table V] The abstract claims that 'prompt-tuning enhances accuracy across various Whisper model sizes, including Small, Medium, and Large,' but Table V shows that for W-S FT the prompt variant degrades WER on Marathi (20.65 to 22.69), Bengali (17.84 to 18.61), and Malayalam (41.79 to 42.37), with an average improvement of only 24.54 to 23.81. This mixed evidence does not support the blanket claim. Please qualify the claim to the languages and model sizes where the improvement is consistent, or provide statistical evidence that the average improvement is reliable.
[All experimental sections] No error bars, number of seeds, or significance tests are reported anywhere in the paper. Given that several headline differences are small (e.g., Table I: W-M FT at 20.15% average versus W-M FT w/ Pro + Tok at 20.05%), the observed differences may be within run-to-run noise. Report standard deviations over multiple seeds or a paired significance test (e.g., bootstrap over utterances) for the central comparisons in Tables I, IV, and V.

minor comments (7)

[Section IV.C] The added BPE tokens are said to be 'derived from datasets emphasizing common sequences in Indian languages,' but the corpus, its size, and the merge algorithm are not specified, making the tokenizer impossible to reproduce. Please disclose these details, ideally by releasing the tokenizer merge file and the vocabulary addition.
[Tables I, IV, and V] The inference-time measurements are reported in minutes without specifying the total audio duration or the number of utterances timed. Please add this information so the reader can interpret the speed numbers.
[Table III] The example sentences for Hindi and Malayalam are not shown in the original script or in a transliteration, so the claims of 27 vs. 19 and 79 vs. 31 tokens cannot be checked. Include the actual sentences and their tokenizations.
[Section V] Hyperparameters such as the optimizer, batch size, learning-rate schedule, warmup, and total number of training steps are not reported. Providing only epochs and gradient accumulation steps is insufficient for replication.
[Section IV.B] It is unclear whether the language-family prompt tokens (<indo>, <dra>) are new learnable embeddings and whether the <SOTP> token is the standard Whisper <|startofprev|> token. Please clarify the initialization and training of these prompt tokens.
[Section VI] The conclusion contains a typo: 'our consistently experiments show' should be 'our experiments consistently show.'
[Notation throughout] The notation W-m with m in {S, M, L} is defined in Section V, but later the paper uses W-S, W-M, and W-L without restating the mapping; consider adding a brief reminder in the table captions.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the WER and inference-time results are empirical measurements; the only concerns (in-sample choice of y=250 and unequal training epochs) are methodological confounds, not circular reasoning.

full rationale

The paper's central claims—that language-family prompt-tuning reduces WER and that a customized BPE tokenizer reduces generated tokens and inference time—are supported by directly measured comparisons on the Kathbath test set (Tables I, IV, V) rather than by quantities that reduce to their own definitions. No load-bearing argument relies on self-citation: the cited related work (e.g., [4], [9], [18]) is external and independent of the authors' claims. Two methodological weaknesses are noted but are not circularity. First, the token count y=250 is selected as the best row of Table II on the same Kathbath test set used to report final WER; this is an in-sample hyperparameter choice that may inflate results, but it is a dataset-selection bias rather than a derivation in which an output equals its input by construction. Second, Section V states that proposed models are fine-tuned for five epochs while 'all the remaining models' run for three; this unequal training budget confounds internal ablations and is a control flaw, but it does not make any reported result equivalent to its input. Overall, no circular step satisfying the quoted-reduction standard was found.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The only invented objects are the prompt tokens <indo> and <dra>, which are categorical design choices rather than falsifiable scientific entities, so they are treated as free parameters and axioms rather than invented entities.

free parameters (2)

y (number of added BPE tokens per language) = 250
Selected because it gives the best average WER in Table II (19.67 vs 21.12 for y=125, 20.20 for y=500, 21.48 for y=1000); this is a test-set-based selection rather than a predefined value.
Number of fine-tuning epochs = 5 for prompt+tokenizer, 3 for others
Reported in Section V; no justification is given for why different epoch counts are used across model variants.

assumptions (4)

domain assumption Kathbath train/validation/test splits from [18] are used unchanged and are representative of the eight languages.
Section III states the same splits as [18] are used; the paper does not report dataset statistics or verify that the test set is not seen during model selection.
domain assumption Language family grouping (Indo-Aryan vs Dravidian) is a valid inductive bias for Whisper's decoder prompting.
Section IV-B introduces <indo>/<dra> prompts based on this assumption; no linguistic analysis or ablation on alternative groupings is provided.
domain assumption Whisper's tokenizer and token head can be extended with new random-weight tokens and fine-tuned without degrading pre-trained performance.
Section IV-C describes the architecture modification; no analysis is given for why random initialization converges or why the original weights are preserved.
ad hoc to paper The BPE tokens are derived from an unspecified Indic text corpus emphasizing common sequences; the paper does not disclose the corpus or the merge algorithm details.
Section IV-C only says 'derived from datasets emphasizing common sequences'; this is a key unstated input to the central method.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing Whisper's Accuracy and Speed for Indian Languages through Prompt-Tuning and Tokenization." pith.science (2026). https://pith.science/paper/7IMPVU4M

@misc{pith2026241219785,
  author       = {Pith},
  title        = {Pith review of: Enhancing Whisper's Accuracy and Speed for Indian Languages through Prompt-Tuning and Tokenization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7IMPVU4M}},
  note         = {Machine review of arXiv:2412.19785}
}

read the original abstract

Automatic speech recognition has recently seen a significant advancement with large foundational models such as Whisper. However, these models often struggle to perform well in low-resource languages, such as Indian languages. This paper explores two novel approaches to enhance Whisper's multilingual speech recognition performance in Indian languages. First, we propose prompt-tuning with language family information, which enhances Whisper's accuracy in linguistically similar languages. Second, we introduce a novel tokenizer that reduces the number of generated tokens, thereby accelerating Whisper's inference speed. Our extensive experiments demonstrate that the tokenizer significantly reduces inference time, while prompt-tuning enhances accuracy across various Whisper model sizes, including Small, Medium, and Large. Together, these techniques achieve a balance between optimal WER and inference speed.

Figures

Figures reproduced from arXiv: 2412.19785 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 15 canonical work pages

[7]

Enhancing multilin- gual speech recognition through language prompt tuning and frame- level language adapter,

S. Li, Y . You, X. Wang, K. Ding, and G. Wan, “Enhancing multilin- gual speech recognition through language prompt tuning and frame- level language adapter,” in ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp. 10 941–10 945

work page 2024
[15]

The Tag- Team Approach: Leveraging CLS and Language Tagging for Enhancing Multilingual ASR,

K. Jayakumar, V . N. Sukhadia, A. Arunkumar, and S. Umesh, “The Tag- Team Approach: Leveraging CLS and Language Tagging for Enhancing Multilingual ASR,” in Proc. INTERSPEECH 2023 , 2023, pp. 4414– 4418

work page 2023
[8]

Improving Whisper's Recognition Performance for Under-Represented Language Kazakh Leveraging Unpaired Speech and Text

J. Li, Y . Pu, Q. Sun, and W.-Q. Zhang, “Improving whisper’s recognition performance for under-represented language kazakh leveraging unpaired speech and text,” arXiv preprint arXiv:2408.05554 , 2024

work page Pith review arXiv 2024
[17]

Improving low-resource languages in pre-trained multilingual language models,

V . Hangya, H. S. Saadi, and A. Fraser, “Improving low-resource languages in pre-trained multilingual language models,” in Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, Y . Goldberg, Z. Kozareva, and Y . Zhang, Eds. Abu Dhabi, United Arab Emirates: Association for Computational Linguistics, Dec. 2022, pp. 11 993–1...

work page 2022
[1]

The dawn of the human-machine era: A forecast of new and emerging language technologies

D. Sayers, R. Sousa-Silva, S. Höhn, L. Ahmedi, K. Allkivi-Metsoja, D. Anastasiou, Š. Be ˇnuš, L. Bowker, E. Bytyçi, A. Catala et al., “The dawn of the human-machine era: A forecast of new and emerging language technologies.” 2021

work page 2021
[2]

Chandramouli and R

C. Chandramouli and R. General, Census of India 2011, Provisional Population Totals. New Delhi: Government of India, 2011

work page 2011
[3]

An overview of indian spoken language recognition from machine learning perspective,

S. Dey, M. Sahidullah, and G. Saha, “An overview of indian spoken language recognition from machine learning perspective,” ACM Trans- actions on Asian and Low-Resource Language Information Processing , vol. 21, no. 6, pp. 1–45, 2022

work page 2022
[4]

Robust speech recognition via large-scale weak supervision,

A. Radford, J. W. Kim, T. Xu, G. Brockman, C. Mcleavey, and I. Sutskever, “Robust speech recognition via large-scale weak supervision,” in Proceedings of the 40th International Conference on Machine Learning , ser. Proceedings of Machine Learning Research, A. Krause, E. Brunskill, K. Cho, B. Engelhardt, S. Sabato, and J. Scarlett, Eds., vol. 202. PMLR, 23...

2023

Show all 21 references

[5]

Scaling speech technology to 1,000+ languages,

V . Pratap, A. Tjandra, B. Shi, P. Tomasello, A. Babu, S. Kundu, A. Elkahky, Z. Ni, A. Vyas, M. Fazel-Zarandi, A. Baevski, Y . Adi, X. Zhang, W.-N. Hsu, A. Conneau, and M. Auli, “Scaling speech technology to 1,000+ languages,” Journal of Machine Learning Research, vol. 25, no....

2024
[6]

Google usm: Scaling automatic speech recognition beyond 100 languages,

Y . Zhang, W. Han, J. Qin, Y . Wang, A. Bapna, Z. Chen, N. Chen, B. Li, V . Axelrod, G. Wang et al. , “Google usm: Scaling automatic speech recognition beyond 100 languages,” arXiv preprint arXiv:2303.01037 , 2023

2023 arXiv
[9]

Vistaar: Diverse Benchmarks and Training Sets for Indian Language ASR,

K. Bhogale, S. Sundaresan, A. Raman, T. Javed, M. M. Khapra, and P. Kumar, “Vistaar: Diverse Benchmarks and Training Sets for Indian Language ASR,” in Proc. INTERSPEECH 2023 , 2023, pp. 4384–4388

2023
[10]

Towards building asr systems for the next billion users,

T. Javed, S. Doddapaneni, A. Raman, K. S. Bhogale, G. Ramesh, A. Kunchukuttan, P. Kumar, and M. M. Khapra, “Towards building asr systems for the next billion users,” in Proceedings of the AAAI Conference on Artificial Intelligence , 2022 (to appear)

2022
[11]

Clsril-23: Cross lingual speech representations for indic languages,

A. Gupta, H. S. Chadha, P. Shah, N. Chhimwal, A. Dhuriya, R. Gaur, and V . Raghavan, “Clsril-23: Cross lingual speech representations for indic languages,” arXiv preprint arXiv:2107.07402 , 2021

2021 arXiv
[12]

A survey of multilingual models for automatic speech recognition,

H. Yadav and S. Sitaram, “A survey of multilingual models for automatic speech recognition,” in Proceedings of the Thirteenth Language Resources and Evaluation Conference , N. Calzolari, F. Béchet, P. Blache, K. Choukri, C. Cieri, T. Declerck, S. Goggi, H. Isahara, B. Maegaard...

2022
[13]

Large language models: A survey,

S. Minaee, T. Mikolov, N. Nikzad, M. Chenaghlu, R. Socher, X. Am- atriain, and J. Gao, “Large language models: A survey,” arXiv preprint arXiv:2402.06196, 2024

2024 arXiv
[14]

Extending whisper with prompt tuning to target-speaker asr,

H. Ma, Z. Peng, M. Shao, J. Li, and J. Liu, “Extending whisper with prompt tuning to target-speaker asr,” in IEEE ICASSP, 2024

2024
[16]

ASR for Low Resource and Multilin- gual Noisy Code-Mixed Speech,

T. Verma, A. Shree, and A. Modi, “ASR for Low Resource and Multilin- gual Noisy Code-Mixed Speech,” in Proc. INTERSPEECH 2023, 2023, pp. 3242–3246

2023
[18]

Indicsuperb: A speech processing universal performance benchmark for indian languages,

T. Javed, K. Bhogale, A. Raman, P. Kumar, A. Kunchukuttan, and M. M. Khapra, “Indicsuperb: A speech processing universal performance benchmark for indian languages,” in Proceedings of the AAAI Confer- ence on Artificial Intelligence , vol. 37, no. 11, 2023, pp. 12 942–12 950

2023
[19]

Language models are unsupervised multitask learners,

A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, and I. Sutskever, “Language models are unsupervised multitask learners,” 2019

2019
[20]

Whisperx: Time- accurate speech transcription of long-form audio,

M. Bain, J. Huh, T. Han, and A. Zisserman, “Whisperx: Time- accurate speech transcription of long-form audio,” arXiv preprint arXiv:2303.00747, 2023

2023 arXiv
[21]

faster-whisper: Faster whisper transcription and translation model,

SYSTRAN, “faster-whisper: Faster whisper transcription and translation model,” https://github.com/SYSTRAN/faster-whisper, 2023, accessed: 2024-08-10

2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [7]

Enhancing multilin- gual speech recognition through language prompt tuning and frame- level language adapter,

S. Li, Y . You, X. Wang, K. Ding, and G. Wan, “Enhancing multilin- gual speech recognition through language prompt tuning and frame- level language adapter,” in ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp. 10 941–10 945

work page 2024

[2] [15]

The Tag- Team Approach: Leveraging CLS and Language Tagging for Enhancing Multilingual ASR,

K. Jayakumar, V . N. Sukhadia, A. Arunkumar, and S. Umesh, “The Tag- Team Approach: Leveraging CLS and Language Tagging for Enhancing Multilingual ASR,” in Proc. INTERSPEECH 2023 , 2023, pp. 4414– 4418

work page 2023

[3] [8]

Improving Whisper's Recognition Performance for Under-Represented Language Kazakh Leveraging Unpaired Speech and Text

J. Li, Y . Pu, Q. Sun, and W.-Q. Zhang, “Improving whisper’s recognition performance for under-represented language kazakh leveraging unpaired speech and text,” arXiv preprint arXiv:2408.05554 , 2024

work page Pith review arXiv 2024

[4] [17]

Improving low-resource languages in pre-trained multilingual language models,

V . Hangya, H. S. Saadi, and A. Fraser, “Improving low-resource languages in pre-trained multilingual language models,” in Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, Y . Goldberg, Z. Kozareva, and Y . Zhang, Eds. Abu Dhabi, United Arab Emirates: Association for Computational Linguistics, Dec. 2022, pp. 11 993–1...

work page 2022

[5] [1]

The dawn of the human-machine era: A forecast of new and emerging language technologies

D. Sayers, R. Sousa-Silva, S. Höhn, L. Ahmedi, K. Allkivi-Metsoja, D. Anastasiou, Š. Be ˇnuš, L. Bowker, E. Bytyçi, A. Catala et al., “The dawn of the human-machine era: A forecast of new and emerging language technologies.” 2021

work page 2021

[6] [2]

Chandramouli and R

C. Chandramouli and R. General, Census of India 2011, Provisional Population Totals. New Delhi: Government of India, 2011

work page 2011

[7] [3]

An overview of indian spoken language recognition from machine learning perspective,

S. Dey, M. Sahidullah, and G. Saha, “An overview of indian spoken language recognition from machine learning perspective,” ACM Trans- actions on Asian and Low-Resource Language Information Processing , vol. 21, no. 6, pp. 1–45, 2022

work page 2022

[8] [4]

Robust speech recognition via large-scale weak supervision,

A. Radford, J. W. Kim, T. Xu, G. Brockman, C. Mcleavey, and I. Sutskever, “Robust speech recognition via large-scale weak supervision,” in Proceedings of the 40th International Conference on Machine Learning , ser. Proceedings of Machine Learning Research, A. Krause, E. Brunskill, K. Cho, B. Engelhardt, S. Sabato, and J. Scarlett, Eds., vol. 202. PMLR, 23...

2023