Pith. sign in

REVIEW 4 major objections 6 minor 34 references

Noise-Agnostic Multitask Whisper Training for Reducing False Alarm Errors in Call-for-Help Detection

T0 review · 4 major / 6 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 By adding a noise-classification head to Whisper's encoder and training it jointly with speech recognition, this paper claims call-for-help detection on real-world recordings improves from 65.03% to 88.48% accuracy while sharply cutting…

desk verdict A plausible application-level trick—adding a noise-classification head to Whisper—but the headline gain is confounded by a gated evaluation that never reports speech-only detection or miss rate. read the letter →

arxiv 2501.11631 v1 pith:BCJZBJL4 submitted 2025-01-20 cs.SD cs.AIeess.AS

Myeonghoon Ryu , June-Woo Kim , Minseok Oh , Suji Lee , Han Park This is my paper · ORCID
classification cs.SDcs.AIeess.AS
keywords call-for-helpdetectionkeywordspottingWhispermultitasklearningnoiseclassificationfalsealarmreductionacousticscenereal-worldrecordings
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 show that a call-for-help detector can be built by fine-tuning an off-the-shelf speech recognition model instead of training a keyword spotter from scratch, and that false alarms in noisy real-world settings can be reduced by teaching the same encoder to also classify environmental noise. The proposed multitask model joins a noise-classification head to Whisper's encoder and trains the noise loss together with the ASR loss. On 547 field recordings, it reaches 88.48% accuracy and 0.89 macro F1, versus 65.03% and 0.63 for Whisper fine-tuned without the noise head. If true, this gives a simple, low-cost way to make emergency voice detection tolerant of background noise, since new keywords only require changing the text-to-label mapping rather than retraining. The paper's own out-of-domain noise test, at 58.82% accuracy, shows this tolerance has limits.

What carries the argument

The machinery is a single shared ASR encoder with two output paths. The usual sequence-to-sequence decoder predicts the transcript, while a second head averages the encoder output over time and predicts one of 13 acoustic-scene classes. Training minimizes the multitask loss $L_{\text{Multi}} = L_{\text{Noise}} + L_{\text{Seq2Seq}}$, a sum of two cross-entropy losses, so the encoder must represent both phonetic content and acoustic environment. At inference, the noise head acts as a gate: if it flags noise, the system stops before decoding; otherwise the transcript is mapped by text to saveme, helpme, or others. This design reuses the pretrained Whisper backbone with only 384 extra hidden units for Whisper-tiny, avoiding a dedicated keyword-spotting head or a separate noise detector.

What would settle it

Run the multitask model on emergency speech mixed with sounds it never saw, especially background human speech, at signal-to-noise ratios well below 20 dB and with different microphones, and count how often the noise head blocks a real call for help or lets a false alarm through; if detection accuracy on such clips falls to the 58.82% noise-classification level or below the fine-tuned-only 65.03%, the claim of noise-agnostic robustness is refuted.

Watch

Extended reading notes

Core claim

The central claim is that a single Whisper encoder can serve two tasks at once, transcribing speech and naming the acoustic scene, and that this joint training changes what the encoder represents, making the downstream call-for-help classifier less trigger-happy in noise. The paper demonstrates this by comparing three systems on its own real-world recordings: vanilla Whisper-tiny at 56.86% accuracy and 0.53 F1, Whisper fine-tuned on emergency speech alone at 65.03% and 0.63, and Whisper with the multitask noise head at 88.48% and 0.89. The confusion matrices show the multitask version converting many false alarms into correct rejections. The paper also reports 98.43% noise-classification accuracy on the in-domain CochlScene test set but only 58.82% on the out-of-domain MS-SNSD set, so the claim is specifically about noise seen during training plus the modest variety in the field recordings.

Load-bearing premise

The load-bearing premise is that the 547 field recordings, 347 speech clips at roughly 20 dB signal-to-noise ratio from eight speakers plus 200 noise clips from a small set of indoor and public environments, represent the acoustic conditions where the detector will be used; the paper's own 58.82% out-of-domain noise accuracy is the visible crack in that premise.

Editorial extensions

If this is right

  • Call-for-help detection can be done by fine-tuning a pretrained ASR model and reading its transcripts, rather than by training a dedicated keyword classifier, which avoids retraining when keywords change.
  • Adding the noise-classification head and multitask objective raises real-recording accuracy from 65.03% to 88.48% and macro F1 from 0.63 to 0.89.
  • The shared encoder with 384 additional units reaches 98.43% noise classification on the in-domain test set, meaning the system can often skip decoding on noise-only inputs and save computation.
  • Because only the text labels saveme, helpme, and others are used after decoding, other emergency phrases can be added to the detection rule without changing model weights.
  • The method's generalization is limited: noise classification drops to 58.82% on out-of-domain MS-SNSD data, so broad real-world noise diversity remains unsolved.

Reading between the lines

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

  • Editorial inference: because classification happens on ASR transcripts, a deployer could add or change keywords by editing the text-to-label mapping, without retraining; the paper demonstrates saveme and helpme but the architecture would support this extension.
  • Editorial inference: the inference-time gate that discards 'noise' clips creates an unmeasured failure mode: if the noise head mistakes urgent speech for noise, the system silently misses a real call-for-help, so deployments should track missed-detection rate, not just false alarms.
  • Editorial inference: the drop to 58.82% on out-of-domain noise suggests an open-set or confidence-thresholded noise model might transfer better to unseen microphones and environments than the fixed 13-scene classifier; this is a testable variant, not a paper claim.
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

4 major / 6 minor

Summary. The paper proposes fine-tuning Whisper-tiny for call-for-help detection with three output classes (saveme, helpme, others) and adds a noise-classification head on the Whisper encoder, trained jointly with ASR via a multitask loss L = L_Noise + L_Seq2Seq. At inference, if the noise head labels the input as noise, processing stops; otherwise ASR transcripts the input and the system classifies it into one of the three call-for-help categories. Experiments compare vanilla Whisper, fine-tuned Whisper, and the multitask variant on 547 field recordings and report 88.48% accuracy / 0.89 macro F1 for multitask versus 65.03% / 0.63 for fine-tuned Whisper. The paper also reports 98.43% in-domain noise classification accuracy and 58.82% out-of-domain accuracy on MS-SNSD.

Significance. The core idea is simple, computationally light, and practically motivated: add a small noise-classification head to a pretrained ASR encoder and train it jointly with ASR, so that the system can reject non-speech noise before running the decoder. If the claimed gains hold under a properly designed evaluation, this would be a useful contribution for emergency call-for-help detection and for KWS systems deployed in noisy acoustic environments. The paper's strengths include leveraging an off-the-shelf Whisper model, adding only a minimal number of parameters, using a real-world recording set, and promising code release. However, the current headline comparison conflates noise rejection with improved call-for-help detection, and the central claims about false-alarm reduction are not directly measured. The result is therefore plausible but not yet established by the evidence presented.

major comments (4)
  1. [Section III-A, Table II, and Section II-C] The headline comparison in Table II mixes 347 speech clips with 200 pure-noise clips and reports only a single 3-class accuracy and macro-F1. Because the proposed inference pipeline in Fig. 2 uses the noise head as a gate—if the input is labeled 'noise', processing terminates—the multitask system can correctly reject all 200 noise clips by construction, while the fine-tuned Whisper baseline has no such gate and may hallucinate 'help me' on those clips. The 23.45-point accuracy gap can therefore be dominated by noise rejection rather than by improved detection of actual calls-for-help. The paper never reports speech-only detection accuracy, per-class recall and precision for saveme/helpme/others, the false-alarm rate on the noise clips, or the miss rate in which a speech clip is incorrectly gated out as noise. These metrics are exactly the ones needed to support the paper's claims of 'reducing false alarms' and 'improved overall call-for-help performance'.
  2. [Section III-B, Table II] All accuracy and F1 results are reported as single runs, without error bars, confidence intervals, or significance tests. Given that Whisper fine-tuning is stochastic and the real-world evaluation set contains only 547 samples, the statement that the multitask method 'significantly outperformed' the fine-tuned baseline is not statistically supported. Please report multiple seeds or at least bootstrap confidence intervals, and state how the reported model was selected.
  3. [Table II and Table III] To attribute the gain to the multitask formulation itself rather than to the mere presence of a noise-rejection function, the comparison should include a fine-tuned Whisper system coupled with a separately trained noise gate (for example, an external VAD or a noise classifier applied to encoder features), as well as the multitask model evaluated without the gate. Without these control arms, the representation-learning benefit of multitask training is not identified separately from the benefit of having any gating mechanism.
  4. [Table III and Section III 'Discussions'] The paper self-reports an out-of-domain noise classification accuracy of only 58.82% on MS-SNSD and acknowledges that the model 'still struggles with noise classification in diverse noisy environments.' Since the deployed system gates out anything labeled noise, this error rate implies a substantial risk that real calls-for-help in untrained noise conditions are silently discarded. This risk is not captured by the aggregate accuracy/F1 on the recording set and should be quantified as a speech miss rate under the proposed gate, ideally across different microphones and SNR conditions, before the paper claims practical deployment suitability.
minor comments (6)
  1. [Abstract] The first sentence is grammatically incomplete: 'keyword spotting is often implemented by keyword classifier to the encoder' should be rephrased, for example as 'keyword spotting is often implemented by attaching a keyword classifier to the encoder of an acoustic model.'
  2. [Table II] There is a typo: 'Vanila Whisper' should be 'Vanilla Whisper'. The baseline names should also be consistent: the table uses both 'Whisper fine-tuned' and the text's 'Whisper fine-tuning alone'.
  3. [Section II-B, Eq. (1)-(3)] The notation is incomplete: the noise label m (and m_i) is not defined, and it is unclear whether the noise classifier predicts binary speech/noise or 13 CochlScene scene classes. Please specify the label space, how the encoder output is pooled over time, and how the averaging omitted in Eqs. (2)-(3) is performed.
  4. [Section III-A, Training details] The statement that 'momentum updates with a coefficient of 0.5 were applied to all learnable parameters' is unusual and needs clarification: does this refer to an exponential moving average of parameters, or to the AdamW momentum? An ablation or a reference would help readers reproduce the training.
  5. [Section III-A, Real-world recording dataset] The paper says the recorded samples are assumed to have an SNR of about 20 dB, but does not explain how this SNR was estimated or controlled. It would be clearer to describe the recording conditions, the microphone positions, and whether any mixing or post-processing was applied.
  6. [Section II-C, Fig. 2] The gating decision is shown as a hard binary 'Noise? Yes -> Ignore' with no threshold or confidence calibration. Since the gate is central to both computational savings and the miss risk, the paper should describe how the noise decision is thresholded and how the threshold affects the trade-off between false alarms and missed calls-for-help.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the multitask result is an empirical comparison on held-out recordings, not a quantity forced by construction.

full rationale

The paper's central claim is empirical: adding a noise-classification head to Whisper and minimizing LMulti = LNoise + LSeq2Seq improves call-for-help detection. The losses are standard cross-entropy terms over external labels—CochlScene acoustic-scene labels for LNoise and Korean emergency ASR text for LSeq2Seq—so no target metric is defined in terms of a fitted parameter. The Table II comparison uses author-recorded samples (347 speech clips and 200 noise clips) that are disjoint from both training corpora, and Table I compares against external models under a shared evaluation protocol. The only self-citations ([15,16]) appear in the discussion as optional future domain-adaptation/augmentation directions and are not load-bearing premises. The acknowledged out-of-domain drop to 58.82% on MS-SNSD is a generalization limitation, not circularity. A legitimate evaluation concern is that the noise-gated inference pipeline can score correct rejections on the 200 pure-noise clips while the no-gate baseline cannot, which could inflate the reported margin; however, that is a measurement-design issue about what Table II's aggregate accuracy means, not a reduction of the result to its own input. No equation, fitted value, or self-citation chain makes the claim equivalent to its assumptions, so no circular step meets the evidentiary bar.

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

No invented entities. The free parameters are standard training hyperparameters chosen by hand. The central assumption is that the recorded evaluation set and CochlScene noise labels represent real deployment conditions; Table III weakens that assumption.

free parameters (3)
  • Momentum coefficient = 0.5
    Chosen by hand for stable multitask training; no selection criterion is stated.
  • Fine-tuning epochs = 10
    Number of epochs chosen without reporting validation-based selection.
  • Learning rate = 1e-4
    Standard AdamW learning rate; no tuning procedure described.
assumptions (5)
  • domain assumption Whisper's pretrained representations transfer to Korean emergency phrases and to the noise classification task.
    The whole method depends on Whisper's encoder features being useful for both ASR and acoustic scene classification after fine-tuning.
  • domain assumption CochlScene acoustic scene labels represent the noise conditions encountered in real-world call-for-help deployments.
    The noise classifier is trained only on CochlScene; Table III shows only 58.82% accuracy on out-of-domain MS-SNSD, weakening this assumption.
  • domain assumption The recorded evaluation samples have an SNR of approximately 20 dB.
    Explicitly stated in Section III-A; if actual SNR differs, the reported accuracy may not hold.
  • ad hoc to paper The six call-for-help scenarios from the Korean emergency dataset can be collapsed into two keywords plus 'others'.
    The paper defines emergency scenarios in Section II-A but does not justify dropping ten of the sixteen dataset scenarios.
  • ad hoc to paper Momentum updates with a coefficient of 0.5 stabilize multitask training.
    A training heuristic not derived or compared with alternatives.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Noise-Agnostic Multitask Whisper Training for Reducing False Alarm Errors in Call-for-Help Detection." pith.science (2026). https://pith.science/paper/BCJZBJL4

@misc{pith2026250111631,
  author       = {Pith},
  title        = {Pith review of: Noise-Agnostic Multitask Whisper Training for Reducing False Alarm Errors in Call-for-Help Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BCJZBJL4}},
  note         = {Machine review of arXiv:2501.11631}
}
read the original abstract

Keyword spotting is often implemented by keyword classifier to the encoder in acoustic models, enabling the classification of predefined or open vocabulary keywords. Although keyword spotting is a crucial task in various applications and can be extended to call-for-help detection in emergencies, however, the previous method often suffers from scalability limitations due to retraining required to introduce new keywords or adapt to changing contexts. We explore a simple yet effective approach that leverages off-the-shelf pretrained ASR models to address these challenges, especially in call-for-help detection scenarios. Furthermore, we observed a substantial increase in false alarms when deploying call-for-help detection system in real-world scenarios due to noise introduced by microphones or different environments. To address this, we propose a novel noise-agnostic multitask learning approach that integrates a noise classification head into the ASR encoder. Our method enhances the model's robustness to noisy environments, leading to a significant reduction in false alarms and improved overall call-for-help performance. Despite the added complexity of multitask learning, our approach is computationally efficient and provides a promising solution for call-for-help detection in real-world scenarios.

Figures

Figures reproduced from arXiv: 2501.11631 by the authors.

Figure 1
Figure 1. Overall illustration of proposed multitask learning [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of the proposed call-for-help detection system using our approach in real-world scenarios. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Confusion matrices of both whisper fine-tuned and multitask approach. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 26 canonical work pages

  1. [1]

    A. Babu, C. Wang, A. Tjandra, K. Lakhotia, Q. Xu, N. Goyal, K. Singh, P. von Platen, Y . Saraf, J. Pino, et al. Xls-r: Self- supervised cross-lingual speech representation learning at scale. arXiv preprint arXiv:2111.09296 , 2021

    arXiv 2021

  2. [2]

    Baevski, Y

    A. Baevski, Y . Zhou, A. Mohamed, and M. Auli. wav2vec 2.0: A framework for self-supervised learning of speech repre- sentations. Advances in neural information processing systems , 33:12449–12460, 2020

    work page 2020

  3. [3]

    A. Berg, M. O’Connor, and M. T. Cruz. Keyword transformer: A self-attention model for keyword spotting. arXiv preprint arXiv:2104.00769, 2021

    arXiv 2021

  4. [4]

    G. Chen, C. Parada, and G. Heigold. Small-footprint keyword spotting using deep neural networks. In 2014 IEEE interna- tional conference on acoustics, speech and signal processing (ICASSP), pages 4087–4091. IEEE, 2014

    work page 2014

  5. [5]

    G. Chen, C. Parada, and T. N. Sainath. Query-by-example keyword spotting using long short-term memory networks. In 2015 IEEE international conference on acoustics, speech and signal processing (ICASSP) , pages 5236–5240. IEEE, 2015

    work page 2015

  6. [6]

    H. Chu, Y . Wang, R. Ju, Y . Jia, H. Wang, M. Li, and Q. Deng. Call for help detection in emergent situations using keyword spotting and paralinguistic analysis. In Companion Publication of the 2021 International Conference on Multimodal Interac- tion, pages 104–111, 2021

    work page 2021

  7. [7]

    Ganin, E

    Y . Ganin, E. Ustinova, H. Ajakan, P. Germain, H. Larochelle, F. Laviolette, M. March, and V . Lempitsky. Domain-adversarial training of neural networks. Journal of machine learning research, 17(59):1–35, 2016

    work page 2016

  8. [8]

    Gong, Y .-A

    Y . Gong, Y .-A. Chung, and J. Glass. AST: Audio Spectrogram Transformer. In Proc. Interspeech 2021, pages 571–575, 2021

    work page 2021

Show all 34 references
  1. [9]

    M. B. Hoy. Alexa, siri, cortana, and more: an introduction to voice assistants. Medical reference services quarterly, 37(1):81– 88, 2018

    2018

  2. [10]

    W.-N. Hsu, B. Bolte, Y .-H. H. Tsai, K. Lakhotia, R. Salakhut- dinov, and A. Mohamed. Hubert: Self-supervised speech representation learning by masked prediction of hidden units. IEEE/ACM Transactions on Audio, Speech, and Language Pro- cessing, 29:3451–3460, 2021

    2021

  3. [11]

    Huang, W

    J. Huang, W. Gharbieh, H. S. Shim, and E. Kim. Query-by- example keyword spotting system using multi-head attention and soft-triple loss. In ICASSP 2021-2021 IEEE Interna- tional Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6858–6862. IEEE, 2021

    2021

  4. [12]

    Hubballi and V

    N. Hubballi and V . Suryanarayanan. False alarm minimization techniques in signature-based intrusion detection systems: A survey. Computer Communications, 49:1–17, 2014

    2014

  5. [13]

    Jeong and J

    I.-Y . Jeong and J. Park. Cochlscene: Acquisition of acoustic scene data using crowdsourcing. In 2022 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), pages 17–21. IEEE, 2022

    2022

  6. [14]

    B. Kim, M. Lee, J. Lee, Y . Kim, and K. Hwang. Query-by- example on-device keyword spotting. In 2019 IEEE automatic speech recognition and understanding workshop (ASRU), pages 532–538. IEEE, 2019

    2019

  7. [15]

    J.-W. Kim, S. Bae, W.-Y . Cho, B. Lee, and H.-Y . Jung. Stethoscope-guided supervised contrastive learning for cross- domain adaptation on respiratory sound classification. In ICASSP 2024-2024 IEEE International Conference on Acous- tics, Speech and Signal Processing (ICASSP),...

    2024

  8. [16]

    J.-W. Kim, M. Toikkanen, S. Bae, M. Kim, and H.-Y . Jung. Repaugment: Input-agnostic representation-level aug- mentation for respiratory sound classification. arXiv preprint arXiv:2405.02996, 2024

    2024 arXiv

  9. [17]

    Kirandevraj, V

    R. Kirandevraj, V . K. Kurmi, V . P. Namboodiri, and C. Jawahar. Generalized keyword spotting using asr embeddings. In 23rd Annual Conference of the International Speech Communica- tion Association, INTERSPEECH 2022 , pages 126–130. ISCA, 2022

    2022

  10. [18]

    Q. Kong, Y . Cao, T. Iqbal, Y . Wang, W. Wang, and M. D. Plumbley. Panns: Large-scale pretrained audio neural networks for audio pattern recognition. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 28:2880–2894, 2020

    2020

  11. [19]

    Z. Liu, T. Li, and P. Zhang. Rnn-t based open-vocabulary keyword spotting in mandarin with multi-level detection. In ICASSP 2021-2021 IEEE International Conference on Acous- tics, Speech and Signal Processing (ICASSP), pages 5649–5653. IEEE, 2021

    2021

  12. [20]

    Mazumder, C

    M. Mazumder, C. Banbury, J. Meyer, P. Warden, and V . J. Reddi. Few-shot keyword spotting in any language. arXiv preprint arXiv:2104.01454, 2021

    2021 arXiv

  13. [21]

    Mo and B

    T. Mo and B. Liu. Encoder-decoder neural architec- ture optimization for keyword spotting. arXiv preprint arXiv:2106.02738, 2021

    2021 arXiv

  14. [22]

    D. Ng, J. Q. Yip, T. Surana, Z. Yang, C. Zhang, Y . Ma, C. Ni, E. S. Chng, and B. Ma. I2cr: Improving noise robustness on keyword spotting using inter-intra contrastive regularization. In 2022 Asia-Pacific Signal and Information Processing Associa- tion Annual Summit and Confe...

    2022

  15. [23]

    Nishu, M

    K. Nishu, M. Cho, and D. Naik. Matching latent encod- ing for audio-text based keyword spotting. arXiv preprint arXiv:2306.05245, 2023

    2023 arXiv

  16. [24]

    D. S. Park, W. Chan, Y . Zhang, C.-C. Chiu, B. Zoph, E. D. Cubuk, and Q. V . Le. Specaugment: A simple data augmentation method for automatic speech recognition. Interspeech 2019 , 2019

    2019

  17. [25]

    Radford, J

    A. Radford, J. W. Kim, T. Xu, G. Brockman, C. McLeavey, and I. Sutskever. Robust speech recognition via large-scale weak supervision. In International conference on machine learning , pages 28492–28518. PMLR, 2023

    2023

  18. [26]

    C. K. Reddy, E. Beyrami, J. Pool, R. Cutler, S. Srinivasan, and J. Gehrke. A scalable noisy speech dataset and online subjective test framework. 2019

    2019

  19. [27]

    Sacchi, A

    N. Sacchi, A. Nanchen, M. Jaggi, and M. Cernak. Open- vocabulary keyword spotting with audio and text embeddings. In INTERSPEECH 2019-IEEE International Conference on Acoustics, Speech, and Signal Processing , 2019

    2019

  20. [28]

    Settle, K

    S. Settle, K. Levin, H. Kamper, and K. Livescu. Query- by-example search with discriminative neural acoustic word embeddings. arXiv preprint arXiv:1706.03818 , 2017

    2017 arXiv

  21. [29]

    H.-K. Shin, H. Han, D. Kim, S.-W. Chung, and H.-G. Kang. Learning audio-text agreement for open-vocabulary keyword spotting. arXiv preprint arXiv:2206.15400 , 2022

    2022 arXiv

  22. [30]

    Tan and Q

    M. Tan and Q. Le. Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning, pages 6105–6114. PMLR, 2019

    2019

  23. [31]

    P. Warden. Speech commands: A dataset for limited-vocabulary speech recognition. arXiv preprint arXiv:1804.03209 , 2018

    2018 arXiv

  24. [32]

    Yang, P.-H

    S.-w. Yang, P.-H. Chi, Y .-S. Chuang, C.-I. J. Lai, K. Lakhotia, Y . Y . Lin, A. T. Liu, J. Shi, X. Chang, G.-T. Lin, et al. Superb: Speech processing universal performance benchmark. arXiv preprint arXiv:2105.01051, 2021

    2021 arXiv

  25. [33]

    Zhang, P

    A. Zhang, P. Zhou, K. Huang, Y . Zou, M. Liu, and L. Xie. U2- kws: Unified two-pass open-vocabulary keyword spotting with keyword bias. In 2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), pages 1–8. IEEE, 2023

    2023

  26. [34]

    Zhang, N

    Y . Zhang, N. Suda, L. Lai, and V . Chandra. Hello edge: Keyword spotting on microcontrollers. arXiv preprint arXiv:1711.07128, 2017

    2017 arXiv

Pith tools

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