Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

FLAM: Frame-Wise Language-Audio Modeling

T0 review · 3 major / 5 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 FLAM casts open-vocabulary sound event detection as frame-wise contrastive classification with event-dependent logit adjustment, and reports large AUROC gains over clip-level baselines while preserving retrieval.

desk verdict FLAM presents a credible empirical advance in open-vocabulary frame-wise audio-text alignment, but the calibration theory in Appendix C.2 relies on an unjustified conditional-independence step and the closed-set evaluation overlaps with training data. read the letter →

arxiv 2505.05335 v2 pith:Z6QLZN2Z submitted 2025-05-08 cs.SD eess.AS

Yusong Wu , Christos Tsirigotis , Ke Chen , Cheng-Zhi Anna Huang , Aaron Courville , Oriol Nieto , Prem Seetharaman , Justin Salamon This is my paper · ORCID
classification cs.SDeess.AS
keywords open-vocabularysoundeventdetectionframe-wisecontrastivelearningaudio-languagemodellogitadjustmenttemporallocalizationsyntheticaudiomixturestext-audioretrievalgrounding
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

FLAM is an audio-language model trained to answer not just what sound is present but when in a clip a described sound occurs. The paper argues that existing contrastive audio-language models align whole clips to captions and therefore cannot pinpoint events, while traditional sound event detectors are locked to a fixed vocabulary. FLAM adds a frame-level contrastive objective that matches each audio frame to text embeddings, with event-dependent logit scales and biases that counter label imbalance and spurious event correlations. To get training labels, the authors synthesize a million 10-second mixtures of labeled events over background audio. If the central claim is right, the result is an open-vocabulary sound event detector with markedly better frame-level AUROC than clip-level baselines, without giving up global retrieval or zero-shot classification.

What carries the argument

The load-bearing object is the frame-wise logit $h(x,l,y)=\alpha_t(y)\,e^{\mathrm{a,loc}}(x)_l \cdot e^t(y)+\beta_t(y)$, trained with a binary cross-entropy objective over all frame-text pairs in a batch. The per-text scale $\alpha_t(y)$ and per-text bias $\beta_t(y)$ come from lightweight MLPs on the text encoder, and the bias MLP is trained on the average per-prompt frame label to approximate the true log-prior ratio. At inference the unbiased score $s\approx\sigma(\log(p(y\mid x,l)/p(y)))$ converts raw frame-text similarity into calibrated detection probabilities. The machinery also includes a chunked ring-based training strategy that computes pairwise frame-text losses across GPUs without collecting all embeddings on one device.

What would settle it

On the actual training mixtures, compute the ratio $p(x,l\mid y,z=1)/p(x,l\mid y,z=-1)$ and compare it with $p(x,l\mid y)/p(x,l)$ for several event types; if the two diverge systematically, the robust-inference derivation in Appendix C.2 breaks. A direct behavioral test would construct a dataset pairing rare short events with frequent long events and check whether the model's 0.5 decision boundary still yields equal detection rates for both event types.

Watch

Extended reading notes

Core claim

The central claim is that open-vocabulary sound event detection can be cast as a per-frame binary contrastive classification problem: for each frame embedding and text query, the model predicts whether the event described by the text is audible at that frame, using the logit $h(x,l,y)=\alpha_t(y)\,e^{\mathrm{a,loc}}(x)_l \cdot e^t(y)+\beta_t(y)$. With a text-dependent logit bias trained by an auxiliary classifier to approximate $\log[p(z=1\mid y)/p(z=-1\mid y)]$, and an unbiased inference score $s=p(z=1\mid x,l,y)/(p(z=1\mid x,l,y)+p(z=1\mid y))\approx\sigma(\log(p(y\mid x,l)/p(y)))$, the paper argues that the model learns calibrated probabilities rather than dataset priors. On a held-out synthetic test set with unseen events it reports AUROC 91.0 versus 74.17 for a retrained clip-level baseline, and on ASFX-SED 81.23 versus 69.56, with retrieval and zero-shot classification approximately preserved.

Load-bearing premise

The derivation of the calibrated classifier assumes that, given a text description, the distribution of audio frames under the positive and negative labels is fully captured by the overall label frequencies and the marginal frame distribution, with no extra interaction between label and frame content; if that conditional-independence step fails, subtracting the learned per-text bias will not produce the claimed unbiased probabilities.

Editorial extensions

If this is right

  • A user can issue an arbitrary textual query for a sound event and receive a temporal boundary for when that event occurs, not just a clip-level match.
  • Closed-set sound event detection benchmarks improve or stay comparable: the paper reports gains on DESED, MAESTRO, AudioSet-Strong, and UrbanSED under AUROC and PSDS metrics.
  • Global retrieval and zero-shot classification do not degrade materially, showing that frame-level supervision and global alignment can be trained jointly.
  • Ablations without the per-text bias and per-text scale show lower F1 across thresholds, indicating that logit adjustment is what makes the outputs calibrated rather than mere frame-level matching.
  • The fixed 10-second input window and coarse 32-frame resolution bound the method; the authors acknowledge this limits longer or more temporally nuanced recordings.

Reading between the lines

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

  • The same recipe of frame-wise binary contrastive loss plus per-category logit adjustment and an unbiased score could transfer to other fine-grained grounding tasks, such as video-text frame grounding or image patch-text segmentation, where positive patches are also rare and imbalanced.
  • The approximation $s\approx\sigma(\log(p(y\mid x,l)/p(y)))$ relies on the per-text bias being strongly negative; in corpora where frequent events have priors closer to balanced, the calibrated boundary may drift and should be audited per event frequency.
  • The synthetic mixture pipeline places events with random overlap but does not deliberately control event co-occurrence statistics; an explicit test varying co-occurrence frequency would clarify whether the logit correction truly removes spurious dependencies or only re-weights marginal label imbalance.
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 / 5 minor

Summary. The paper proposes FLAM, a contrastive audio-language model that augments a standard CLIP-style global objective with a frame-wise binary classification objective for open-vocabulary sound event detection. The model outputs both a global audio embedding and a temporal sequence of frame embeddings; each frame-text pair is scored by a logit that includes a per-text scale and a per-text bias, trained with a logit-adjustment motivation to handle label imbalance. The model is trained on a large synthetic dataset of 1M mixtures constructed by inserting text-labeled sound events into background audio, together with closed-set SED datasets. The paper reports large gains over a CLAP-style baseline and a retrained MGA-CLAP on held-out synthetic and external benchmarks, while showing comparable performance on retrieval and zero-shot classification, and presents a theoretical appendix claiming that the per-text bias yields a Bayes-optimal unbiased classifier.

Significance. If the empirical results are valid, FLAM is a useful contribution to open-vocabulary audio event detection: it demonstrates that explicit frame-level supervision, together with per-text logit scaling and biasing, can give substantial gains in temporal localization over clip-level contrastive models, and the synthetic data pipeline plus the released ASFX-SED benchmark are valuable resources. The held-out synthetic evaluations (Held-out, ASFX-SED) are well designed and provide credible evidence for the core localization claim. The ablations on global loss, initialization, and temporal granularity are informative. However, the paper's theoretical claims about calibration and unbiasedness rest on an unjustified equality in Appendix C.2, and the closed-set evaluation protocol is not documented; these issues must be resolved before the significance of the claims can be fully assessed.

major comments (3)
  1. [Appendix C.2, Eq. (17)] The derivation silently replaces log[pdata(x,l|y,z=1)/pdata(x,l|y,z=-1)] with log[pdata(x,l|y)/pdata(x,l)] in the expression for pdata(z=1|x,l,y). This equality is not a consequence of Bayes rule and is generally false under the generative model of §4.2: the left-hand ratio is the per-frame likelihood ratio for event presence given the text y, while the right-hand ratio marginalizes over z and loses the event-presence information. Since Eq. (9), the approximation in Eq. (8), and the 'unbiased classifier' in Eqs. (7) and (20)-(22) all depend on this step, the claimed Bayes-optimality and calibration guarantees are not established. Please provide a correct derivation, or state and justify an explicit assumption, noting that the natural conditional-independence assumption pdata(x,l|y,z)=pdata(x,l|y) contradicts the detection task itself; alternatively, present the logit adjustment as a heuristic and support it with the empirical F1 results rather than a theoretical proof.
  2. [Section 5.1 / 5.2, Table 1] The paper states that FLAM is trained on AudioSet-Strong, DESED, and UrbanSED (Section 5.1, Appendix C.6), but Table 1 reports SED results on these same datasets without specifying the evaluation split. If the model was trained on the same recordings used for evaluation, the closed-set AUROC/PSDS numbers are in-sample and do not support the abstract's claim of outperforming prior models on closed-set SED. Please specify the exact train/validation/test partitions used for each dataset and confirm that no evaluation clips were included in training. This is essential for interpreting the closed-set results in Table 1 and the ablation in Table 4.
  3. [Section 5.3] The claim that FLAM produces 'accurate and calibrated probabilities' and that the unbiased classifier is 'calibrated' is not supported by quantitative calibration metrics. The only quantitative evidence is the F1/precision/recall curve in Figure 3 and qualitative examples in Figures 2 and 4-9. Since the theoretical calibration argument in Appendix C.2/Eq. (17) is not established, the empirical calibration claim needs direct evidence. Please include reliability diagrams or expected calibration error on held-out data, or revise the wording to 'well-behaved scores' rather than 'calibrated probabilities'.
minor comments (5)
  1. [Table 1] The table formatting is broken (e.g., '9.3791.66' and '29.52 93.62' run together), and the bold-number convention is unclear. In the DESED PSDS column, MGA-CLAP (reported) has 26.4, which is higher than FLAM's 9.37, so the table should not imply FLAM is best on every closed-set metric; the text acknowledges the DESED exception, but the table should be made consistent.
  2. [Appendix C.5 / C.6] Please state explicitly that the Clotho and AudioCaps training uses only the official training splits and that evaluation in Table 2 is on the official test splits. The current wording 'We sample data from our (1.1M) dataset, AudioCaps, and Clotho' leaves open the possibility of train/test overlap.
  3. [Appendix C.3, Eq. (19)] The notation '∝_z' is confusing. Please define it as 'proportional as a function of z' or replace it with a clearer equivalence relation to avoid ambiguity.
  4. [Section 3.3] The ring-based chunked loss computation is described at a high level. Please provide pseudocode or a precise description of how masks are used and how the per-GPU loss terms are accumulated, so that the memory-efficient strategy is reproducible.
  5. [Section 3.2 and Contribution List] The term 'unbiased classifier' is potentially misleading in light of the derivation gap. Consider renaming it to 'bias-corrected classifier' throughout the paper.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: FLAM's frame-wise SED claims are tested against held-out synthetic and external benchmarks, and the logit-adjusted objective is a trained component rather than a disguised prediction.

full rationale

The central claims are evaluated on held-out synthetic mixtures (unseen events and backgrounds), on ASFX-SED built from an external sound-effects library, and on standard closed-set SED, retrieval, and zero-shot benchmarks, so the reported gains are not an echo of the training objective. The per-text logit scale and bias are trainable components fitted to the SED objective and a prior loss; they are not presented as independent predictions, and the paper does not rename fitted values as results. Self-citations (e.g., Tsirigotis et al. 2023, Wu* et al. 2023, Chen et al. 2022) are background method or architecture references and are not load-bearing: the Bayes-optimal classifier is derived in Appendices C.2-C.3 rather than imported as an unverified theorem from the authors' own work. One non-circular correctness caveat should be flagged: Appendix C.2 Eq. 17 replaces log[pdata(x,l|y,z=1)/pdata(x,l|y,z=-1)] with log[pdata(x,l|y)/pdata(x,l)] without stating a conditional-independence or balance condition; if this equality fails, the theoretical guarantee for Eq. 9 is unsupported. This is a proof gap affecting the derivation's validity, not a reduction of a prediction to its inputs, so it does not make the empirical evaluation circular.

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

The central derivation rests on a Bayes-optimum argument that needs one unstated conditional-independence equality (Eq. 17, C.2), plus a test-time invariance assumption (C.3), and on the representativeness of synthetic mixtures. The per-text scale and bias are learned functions fitted during training. No new physical entities are introduced.

free parameters (4)
  • Per-text logit scale alpha_t(y) = Learned MLP on text embeddings; initialized at log(10)
    Controls per-event sharpness of frame-text logits; trained with LSED. Central to the calibrated detector claim.
  • Per-text logit bias beta_t(y) = Learned MLP on text embeddings; initialized at -8
    Intended to absorb label imbalance per event; trained with Lp on batch average labels. Subtracting it gives the robust classifier in the paper's derivation.
  • Loss weights gamma_SED, gamma_p = 200, 1
    Chosen by hand so LSED and LCLIP have comparable scale and Lp converges; affects the balance between frame-level and global objectives.
  • Synthesis hyperparameters = N~U(1,10), max 3 overlapping events, offset U(6,30) dB, split/repeat 10% each
    Define the simulated data distribution used for all frame-level supervision; chosen by design rather than fitted to benchmarks.
assumptions (5)
  • standard math Bayes-rule decomposition and logistic-regression optimum (Appendix C.1-C.3)
    Used to derive the SigLIP bias interpretation and the claimed optimum of LSED.
  • ad hoc to paper Eq. 17 conditional-independence equality: p(x,l|y,z=1)/p(x,l|y,z=-1) = p(x,l|y)/p(x,l)
    Without this unstated equality, h*=log(p(y|x,l)/p(y))+beta*(y) does not follow; the paper does not justify it.
  • domain assumption Test-time invariance: pdata(x,l|y,z)=ptest(x,l|y,z) and ptest(z=1|y)=0.5
    Needed for the robust classifier in C.3; plausible but unverified, especially for real-world deployment.
  • domain assumption Unlabeled frames are negative for event descriptions not listed for that clip (Sec 3.1)
    Defines the negative sampling in LSED; if a caption happens to describe an unlisted sound, labels are noisy.
  • domain assumption Synthetic mixtures are representative of open-vocabulary sound event distributions (Sec 4.2)
    The main evidence for open-vocabulary localization comes from test sets built with the same simulation pipeline; real-world generalization is not directly measured.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FLAM: Frame-Wise Language-Audio Modeling." pith.science (2026). https://pith.science/paper/Z6QLZN2Z

@misc{pith2026250505335,
  author       = {Pith},
  title        = {Pith review of: FLAM: Frame-Wise Language-Audio Modeling},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Z6QLZN2Z}},
  note         = {Machine review of arXiv:2505.05335}
}
read the original abstract

Recent multi-modal audio-language models (ALMs) excel at text-audio retrieval but struggle with frame-wise audio understanding. Prior works use temporal-aware labels or unsupervised training to improve frame-wise capabilities, but they still lack fine-grained labeling capability to pinpoint when an event occurs. While traditional sound event detection models can precisely localize events, they are limited to pre-defined categories, making them ineffective for real-world scenarios with out-of-distribution events. In this work, we introduce FLAM, an open-vocabulary contrastive audio-language model capable of localizing specific sound events. FLAM employs a memory-efficient and calibrated frame-wise objective with logit adjustment to address spurious correlations, such as event dependencies and label imbalances during training. To enable frame-wise supervision, we leverage a large-scale dataset with diverse audio events, LLM-generated captions and simulation. Experimental results and case studies demonstrate that FLAM significantly improves the open-vocabulary localization capability while maintaining strong performance in global retrieval and downstream tasks.

Figures

Figures reproduced from arXiv: 2505.05335 by the authors.

Figure 1
Figure 1. Comparison of traditional audio-language contrastive modeling (left) with the proposed Frame-wise Language-Audio Modeling (FLAM, right). Traditional models generate global embeddings for audio-text pairs and optimize an instance-wise contrastive objective. In contrast, FLAM integrates both instance-wise and fine-grained frame-wise contrastive objectives. Audio mixtures are synthesized from event instances, paired wi… view at source ↗
Figure 2
Figure 2. Example outputs from FLAM and MGA-CLAP on open-vocabulary sound-event-detection example (top row) with detailed value at each frame (bottom row). FLAM, trained with event-dependent logit adjustment, converts raw cosine similarity into calibrated predictions. By contrast, the unsupervised MGA-CLAP model produces less accurate results and cannot be calibrated from its output. a batch BSED randomly sampled from the tra… view at source ↗
Figure 3
Figure 3. Framewise F1, precision and recall on ASFX-SED dataset. FLAM achieves higher F1 under various threshold com￾pared to ablated models without text-dependent bias and scale, showing the effectiveness of our logit-correction objective. with the threshold, indicating that FLAM’s output is polar￾ized due to the unbiased classifier transforming raw outputs into near-binary probabilities. Additional qualitative results in §… view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Sound event detection results of FLAM on ASFX-SED dataset 12 [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Sound event detection results of FLAM on synthetic held-out dataset. 13 [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]
Figure 6
Figure 6. Figure 6: Sound event detection results of FLAM on Audioset-Strong dataset. 14 [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Sound event detection results of FLAM on ASFX-SED dataset compared with ablation models. 15 [PITH_FULL_IMAGE:figures/full_fig_p015_7.png]
Figure 8
Figure 8. Figure 8: Sound event detection results of FLAM on synthetic held-out dataset compared with ablation models. C. Appendix C.1. SigLIP Objective SigLIP (Zhai et al., 2023) treats contrastive alignment as a binary classification problem LSigLIP = − 1 B X B i=1 X B j=1 log σ [PITH_…
Figure 9
Figure 9. Figure 9: Sound event detection results of FLAM on Audioset-Strong dataset compared with ablation models. the logit adjustment for the label-imbalanced binary classification problem, which ultimately leads to the formulation of the objective in §3. Consider the binary classifica…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Melody-Lyrics Matching with Contrastive Alignment Loss

    eess.AS 2025-07 unverdicted novelty 6.0 of 10

    The paper proposes a self-supervised contrastive framework with a syllable-level phoneme-stress representation, sylphone, for matching symbolic melodies to lyrics.

  2. Auditory Intelligence: Understanding the World Through Sound

    eess.AS 2025-08 conditional novelty 4.0 of 10

    A position paper proposing four cognitively inspired audio tasks (ASPIRE, SODA, AUX, AUGMENT) to push machine hearing beyond recognition toward explanation, reasoning, and interaction.

Reference graph

Works this paper leans on

20 extracted references · 9 canonical work pages · cited by 2 Pith papers

  1. [1]

    keyword, tag

    = 0.5 . Finally, we consider that the mechanism, that generates positive or negative frames given a certain event, is invariant between train-time and test-time conditions,p data(x,l|y,z) =p test(x,l|y,z). Following Menon et al. (2021) and Tsirigotis et al. (2023), the Bayes-optimal robust classifier is given by z∗(x,l,y) = arg max z∈{−1,1} ptest(z|x,l,y)...

    work page 2021

  2. [6]

    Clap learning audio concepts from natural language su- pervision

    Elizalde, B., Deshmukh, S., Al Ismail, M., and Wang, H. Clap learning audio concepts from natural language su- pervision. InICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1–5. IEEE,

    2023

  3. [7]

    P., Fonseca, E., Jansen, A., Liu, C., Moore, R

    Hershey, S., Ellis, D. P., Fonseca, E., Jansen, A., Liu, C., Moore, R. C., and Plakal, M. The benefit of temporally- strong labels in audio event classification. InICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 366–370. IEEE,

    work page 2021

  4. [9]

    Q., Sablayrolles, A., Roux, A., Mensch, A., Savary, B., Bamford, C., Chaplot, D

    Jiang, A. Q., Sablayrolles, A., Roux, A., Mensch, A., Savary, B., Bamford, C., Chaplot, D. S., Casas, D. d. l., Hanna, E. B., Bressand, F., et al. Mixtral of experts.arXiv preprint arXiv:2401.04088,

    arXiv

  5. [10]

    D., Kim, B., Lee, H., and Kim, G

    Kim, C. D., Kim, B., Lee, H., and Kim, G. Audiocaps: Gen- erating captions for audios in the wild. InProceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Hu- man Language Technologies, Volume 1 (Long and Short Papers), pp. 119–132,

    2019

  6. [11]

    Avoiding spurious correlations via logit correction

    Liu, S., Zhang, X., Sekhar, N., Wu, Y ., Singhal, P., and Fernandez-Granda, C. Avoiding spurious correlations via logit correction. InThe Eleventh International Confer- ence on Learning Representations, 2023b. Liu, Y . Roberta: A robustly optimized bert pretraining approach.arXiv preprint arXiv:1907.11692, 364,

    arXiv 1907

  7. [12]

    Hierarchical text-conditional image generation with clip latents.arXiv preprint arXiv:2204.06125, 1(2):3,

    Ramesh, A., Dhariwal, P., Nichol, A., Chu, C., and Chen, M. Hierarchical text-conditional image generation with clip latents.arXiv preprint arXiv:2204.06125, 1(2):3,

    arXiv

  8. [13]

    Sound event detection in synthetic domestic environments

    Serizel, R., Turpault, N., Shah, A., and Salamon, J. Sound event detection in synthetic domestic environments. In ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 86–90. IEEE,

    work page 2020

Show all 20 references
  1. [15]

    P., and Salamon, J

    Wu, H.-H., Nieto, O., Bello, J. P., and Salamon, J. Audio- text models do not yet leverage natural language. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1–5. IEEE,

    2023

  2. [16]

    Large-scale contrastive language- audio pretraining with feature fusion and keyword-to- caption augmentation

    Wu*, Y ., Chen*, K., Zhang*, T., Hui*, Y ., Berg-Kirkpatrick, T., and Dubnov, S. Large-scale contrastive language- audio pretraining with feature fusion and keyword-to- caption augmentation. InICASSP 2023-2023 IEEE In- ternational Conference on Acoustics, Speech and Signal Pro...

    2023

  3. [17]

    Towards weakly supervised text-to-audio grounding.arXiv preprint arXiv:2401.02584,

    Xu, X., Ma, Z., Wu, M., and Yu, K. Towards weakly supervised text-to-audio grounding.arXiv preprint arXiv:2401.02584,

    arXiv

  4. [18]

    D., and Wang, W

    Yuan, Y ., Chen, Z., Liu, X., Liu, H., Xu, X., Jia, D., Chen, Y ., Plumbley, M. D., and Wang, W. T-clap: Temporal- enhanced contrastive language-audio pretraining.arXiv preprint arXiv:2404.17806,

    arXiv

  5. [20]

    and SigLip (Zhai et al., 2023). C.7. Details of SED Metrics Following MGA-CLAP (Li et al., 2024), we apply a median filter of size 3 frames to the SED output before evaluation. We usesed scores eval(Ebbers et al.,

    2023

  6. [2018]

    Clotho: An audio 9 FLAM: Frame-Wise Language-Audio Modeling captioning dataset

    Drossos, K., Lipping, S., and Virtanen, T. Clotho: An audio 9 FLAM: Frame-Wise Language-Audio Modeling captioning dataset. InICASSP 2020-2020 IEEE Inter- national Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 736–740. IEEE,

    2020

  7. [2019]

    Mean teacher convolution system for dcase 2018 task

    JiaKai, L. Mean teacher convolution system for dcase 2018 task

    2018

  8. [2020]

    Threshold independent evaluation of sound event detection scores

    Ebbers, J., Haeb-Umbach, R., and Serizel, R. Threshold independent evaluation of sound event detection scores. InICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1021–1025. IEEE,

    2022

  9. [2021]

    Hts-at: A hierarchical token-semantic audio transformer for sound classification and detection

    Chen, K., Du, X., Zhu, B., Ma, Z., Berg-Kirkpatrick, T., and Dubnov, S. Hts-at: A hierarchical token-semantic audio transformer for sound classification and detection. InICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 646...

    2022

  10. [2022]

    Dcase 2024 task 4: Sound event detection with heterogeneous data and missing labels.arXiv preprint arXiv:2406.08056,

    Cornell, S., Ebbers, J., Douwes, C., Mart ´ın-Morat´o, I., Harju, M., Mesaros, A., and Serizel, R. Dcase 2024 task 4: Sound event detection with heterogeneous data and missing labels.arXiv preprint arXiv:2406.08056,

    2024 arXiv

  11. [2023]

    Representa- tion learning with contrastive predictive coding.ArXiv, abs/1807.03748,

    van den Oord, A., Li, Y ., and Vinyals, O. Representa- tion learning with contrastive predictive coding.ArXiv, abs/1807.03748,

    arXiv

  12. [2024]

    Bert: Pre-training of deep bidirectional trans- formers for language understanding.arXiv preprint arXiv:1810.04805,

    Devlin, J. Bert: Pre-training of deep bidirectional trans- formers for language understanding.arXiv preprint arXiv:1810.04805,

    arXiv

Pith tools

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