Enhancing Stereo Sound Event Detection with BiMamba and Pretrained PSELDnet

Han Yin; Wenmiao Gao

REVIEW 3 major objections 6 minor 34 references

Enhancing Stereo Sound Event Detection with BiMamba and Pretrained PSELDnet

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Replacing the Conformer decoder of a pretrained PSELDnet with a bidirectional Mamba block and asymmetric convolutions yields the best stereo sound event localization and detection scores on the DCASE2025 development set while using 76M…

desk verdict Useful efficiency-oriented tweak on PSELDnet for stereo SELD, but the headline accuracy gain rests on single-run dev-set scores with no variance estimates. read the letter →

arxiv 2507.09570 v1 pith:HUQ3TFZV submitted 2025-07-13 eess.AS cs.SD

Wenmiao Gao , Han Yin This is my paper

classification eess.AScs.SD

keywords stereosoundeventlocalizationanddetectionBiMambastatespacemodelsPSELDnetpretrainingasymmetricconvolutionDCASE2025Task3efficientaudioMulti-ACCDOA

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 claims that a stereo sound-event localization and detection system can be made both more accurate and far cheaper by swapping the heavy Conformer decoder of a pretrained PSELDnet for a bidirectional Mamba block augmented with asymmetric convolutions. On the DCASE2025 Task 3 development set, the resulting BiMambaAC model reaches 39.6% location-dependent F1 at a 20-degree threshold and 15.8 degrees DOA error using 76M parameters, beating the Conformer baseline (38.2% F20, 210M parameters) and the plain BiMamba variant (36.2% F20, 178M parameters). The paper also reports that the same model placed 9th in the DCASE 2025 Task 3 challenge with a single model, no external data, and only simple channel-swapping augmentation. The practical stakes are efficiency: state-space sequence models offer a path to competitive SELD without Transformer-scale compute.

What carries the argument

The load-bearing component is the BiMamba2DAC block: a bidirectional state-space sequence layer (two Mamba directions over time) combined with an asymmetric convolution pathway that applies 1D convolutions separately over the time axis and the frequency axis instead of a single 2D convolution. This block replaces the Conformer decoder in the pretrained PSELDnet pipeline, while a temporal module restores and aligns the temporal resolution after CNN14 pooling. The argument is carried by two mechanisms: state-space selectivity lets the model track overlapping events without quadratic attention cost, and time-frequency decoupling extracts complementary features at lower compute.

What would settle it

Run BiMambaAC and the Conformer baseline 10 times each with different seeds on the DCASE2025 Task 3 development set, selecting checkpoints by validation F1 as the paper does, and compare the mean and standard deviation of F20; the claimed advantage is unsupported if the Conformer mean reaches or exceeds the BiMambaAC mean within one standard deviation.

Watch

Extended reading notes

Core claim

The paper claims that, in the fine-tuning scenario, its proposed BiMambaAC model — a pretrained CNN14 encoder with a bidirectional Mamba decoder augmented by asymmetric convolution operating separately on time and frequency — is the best overall system on the DCASE2025 Task 3 stereo SELD development set. It reports 39.6% location-dependent F1 at a 20° threshold (F20) and 15.8° DOA error with 76M parameters, compared with 38.2% F20 for the original Conformer model at 210M parameters, 36.2% F20 for plain BiMamba at 178M parameters, and 35.1% F20 for HTS-AT at 28M parameters. The ablation attributes an 8.6-point F20 drop to removing bidirectional processing and a 6.4-point drop to removing asymmetric convolution, and the paper notes that increasing the state dimension from 64 to 128 adds only 0.3% F20 while costing 6GB of extra GPU memory. On the official challenge evaluation dataset the same single-model system scores 31.0% F20, which the paper reports as 9th place with no external data and only simple channel-swapping augmentation.

Load-bearing premise

The comparison assumes that a single development-set run per model, with the checkpoint chosen by validation F1 and no significance testing, is stable enough to support the 1.4-point F20 advantage over the Conformer baseline.

Editorial extensions

If this is right

Fine-tuning from pretrained PSELDnet weights helps every architecture in the comparison, and BiMambaAC is the top performer on the development set (39.6% F20, 15.8° DOAE).
Both new components matter: taking away bidirectional processing costs 8.6 points of F20, and taking away asymmetric convolution costs 6.4 points.
Distilling the Conformer's heavy decoder into a 76M-parameter BiMamba decoder preserves or improves accuracy while reducing parameters by more than half and MACs by 39% relative to the plain BiMamba variant.
The efficiency gain is operational: the default dstate=64 configuration is preferred over dstate=128 because the larger state adds only 0.3% F20 but 6GB of training-time GPU memory.
A single-model system with no external training data and only channel swapping can reach 31.0% F20 on the official evaluation set, ahead of some ensemble submissions that rely on external data.

Reading between the lines

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

An untested extension of the same recipe would attach the BiMamba decoder to a different pretrained encoder, such as HTS-AT; the paper's pretraining results suggest that encoder also benefits strongly from pretraining, which would separate encoder gains from decoder gains.
Because the stereo-to-pseudo-FOA conversion zeroes the X and Z components, the model is only estimating left-right direction; learning to predict the missing components from stereo cues, or adding microphone-array geometry, is a natural way to extend the system toward full 3D localization.
The asymmetric-convolution block is a generic time-frequency decoupling module, so the same swap could be tested in single-channel sound event detection, where the compute savings would be even more visible.

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.

Desk Editor's Note

read the letter

Quick take: the paper does something sensible and reports it honestly. The new bits are the specific combination of pretrained PSELDnet with a BiMamba decoder that uses asymmetric convolutions, and the pseudo-FOA stereo adaptation. The components themselves are not new—BiMamba already replaces Conformer in SELD-Mamba—but the combination and the measurements on the DCASE2025 stereo task are not in the cited literature.

What it does well: the ablation is internally consistent, the parameter and MACs reporting is useful, and the comparison set is relevant. The paper does not oversell; its conclusions match its tables. The pseudo-FOA conversion is stated as an engineering assumption, not hidden.

The soft spots are mostly about evidence quality. The central comparison—39.6% vs 38.2% F20 over Conformer in Table II—comes from one checkpoint per model, chosen by validation F1, with no seeds, repeated runs, or confidence intervals. On a 13-class SELD task, a 1.4-point difference is within plausible run-to-run variation. That concern becomes sharper when you look at Table IV: the same BiMambaAC system gets 31.0% F20 on the official evaluation set, an 8.6-point drop that the text neither explains nor analyzes. That gap does not disprove the architecture story, but it does mean the dev-set gain might be partly selection noise.

The ablation drops are also from single runs. Removing asymmetric convolution costs 6.4 points; removing bidirectional Mamba costs 8.6 points. Those are large effects but unquantified. A serious revision should add repeated runs with error bars and some analysis of the dev-to-eval gap.

Minor issues: the GitHub URL as printed has spaces and is not directly usable, and the pseudo-FOA assumption (X and Z set to zero) is unvalidated against alternatives. The self-citations are not load-bearing and are relevant background, so I do not see a citation problem.

Who this is for: anyone working on efficient SELD architectures, especially in the DCASE challenge space. It is a modest, useful data point, not a breakthrough.

Recommendation: it deserves a serious referee. I would send it to peer review, but ask for repeated runs, variance estimates, and a discussion of the generalization gap before acceptance.

Referee Report

3 major / 6 minor

Summary. The paper proposes a stereo sound event localization and detection (SELD) system for DCASE2025 Task 3, built by replacing the Conformer decoder of the pre-trained PSELDnet with a bidirectional Mamba (BiMamba) module augmented by asymmetric convolutions (BiMambaAC). Stereo input is converted to a pseudo-FOA representation via a mid/side-style transformation. The authors evaluate the model on the DCASE2025 development set and the official challenge evaluation, reporting location-dependent F1 (F20), DOA error, relative distance error, parameter count, and MACs. In the fine-tuning scenario, BiMambaAC reaches 39.6% F20 with 76M parameters, outperforming the Conformer-based PSELDnet (38.2% F20, 210M parameters) and BiMamba (36.2% F20, 178M parameters). An ablation study shows drops of 8.6 points without bidirectional processing and 6.4 points without asymmetric convolution. The official evaluation result for the proposed system is 31.0% F20, ranking 9th in the challenge.

Significance. If the reported gains are reproducible, the paper makes a useful contribution: it demonstrates that a Mamba-based decoder can replace a Conformer decoder in a pre-trained SELD system with a favorable accuracy-efficiency trade-off, while the public code release supports reproducibility. The ablation study is internally consistent, and the comparison includes several relevant baselines. However, the central quantitative claim rests on single-run development-set scores without variance estimates, and the large unexplained gap between the development-set F20 (39.6%) and the official evaluation F20 (31.0%) weakens confidence in the headline improvement. The significance is therefore conditional on additional evidence of stability and on a credible explanation of the generalization gap.

major comments (3)

[Section IV-B, Tables II and IV] The central claim that BiMambaAC outperforms BiMamba by +3.4% F20 and the Conformer baseline by +1.4% F20 is supported only by a single development-set run per model, with checkpoint selection based on validation F1 and no repeated seeds, confidence intervals, or significance tests. This is a load-bearing issue because 1–2 point F20 differences are plausibly within run-to-run variance for a 13-class SELD task. Moreover, the same system achieves 39.6% F20 on the development set (Table II) but only 31.0% F20 on the official evaluation (Table IV), an 8.6-point drop that the text neither explains nor analyzes. The authors should report multiple runs with variance, and either analyze the development-to-evaluation gap or temper the accuracy claims accordingly.
[Section IV-A, Table I vs. Abstract] The abstract states that the proposed method performs better than both the baseline and the original PSELDnet with a Conformer decoder, but this is not true in the from-scratch scenario: Table I shows Conformer at 32.7% F20 versus BiMambaAC at 32.1% F20. The improvement over Conformer only holds in the fine-tuning scenario (Table II). The claim should be explicitly qualified to the fine-tuning setting, and the from-scratch result should be discussed rather than ignored.
[Section IV-B, Table III and Table IV] The ablation result without bidirectional Mamba is 31.0% F20, which is numerically identical to the official-evaluation F20 of the proposed system in Table IV. This coincidence is not addressed. It is important to clarify which configuration was actually submitted to the challenge: if the official submission was the full BiMambaAC model, the equality is coincidental and should be noted; if the official submission differed, the paper should state so explicitly. Otherwise, the reader cannot assess whether the challenge result reflects the proposed architecture.

minor comments (6)

[Section II-A, Eq. (2)] The output equation is written as y(t) = C^T h'(t) + Dx(t), but the standard SSM output is y(t) = C^T h(t) + Dx(t). As written, the output depends on the time derivative of the state, which is inconsistent with the surrounding derivation.
[Section II-A, Eq. (3)] The expression for \bar{B} uses (\Delta A)^{-1}, which is not well-defined when \Delta A is singular. The zero-order-hold discretization is usually given in integral or series form; the authors should provide the nonsingular-free formulation.
[Section III-C] The hyperparameters are not fixed across models: batch size ranges from 32 to 256 and weight decay from 1e-4 to 5e-6 depending on model size. This makes the architecture comparisons less controlled; the authors should state whether the same tuning procedure was applied to all baselines or acknowledge the potential confound.
[Table III] The heading contains a typo: 'Asymtric' should be 'Asymmetric'.
[Abstract and Section V] The GitHub URL contains spaces ('https://github.com/ alexandergwm/DCASE2025 TASK3 Stereo PSELD Mamba'); the link should be corrected and verified.
[References] Several references in the related work are by the same author group ([4]–[7], [9], [21], [22]); including them is fine, but the authors may wish to cite additional independent works on Mamba-based SELD to broaden the context.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the claimed gains are measured against the external DCASE2025 benchmark, not derived from fitted inputs or load-bearing self-citations.

full rationale

The paper's central claim is an empirical benchmark: replacing the Conformer decoder with BiMamba plus asymmetric convolution in a pretrained PSELDnet and measuring F20/DOAE/RDE on the DCASE2025 Task 3 development and official evaluation sets. No step in the paper reduces a reported number to a fitted parameter or to a definition. The pseudo-FOA conversion in Eq. (1) sets X(n)=0 and Z(n)=0, explicitly acknowledging that stereo signals inherently lack precise spatial information; the paper presents this transform only as a compatibility mechanism for enabling smooth adaptation to existing pre-trained models, not as a recovery of true FOA or as evidence for the architecture's accuracy. Checkpoint selection by highest validation location-dependent F1-score (Section III-C) and the lack of repeated seeds is a statistical robustness concern, not circularity, because the reported F20 values are independent measurements against an external challenge dataset. The few co-authored citations (e.g., refs. [4] and [7] include author H. Yin) appear only in the introduction as background on sound event detection and domain generalization and are not load-bearing for the BiMambaAC result; no uniqueness theorem or ansatz is imported from the authors' prior work. Since the performance numbers are evaluated on the public DCASE2025 dataset and the pretrained PSELDnet weights are external, the paper is self-contained as an experimental comparison, and the derivation chain does not feed back into its own inputs.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

This is an empirical benchmark paper, so the ledger contains architecture hyperparameters and transfer assumptions rather than fitted physical constants. No free parameter is fit to the headline F20, but the pseudo-FOA transfer assumption and the single-run evaluation design are the largest unproven inputs to the claim.

free parameters (6)

dstate = 64
State dimension of the Mamba blocks, described as a recommended parameter in Section III-C. The ablation shows dstate=128 gives only +0.3% F20 at 6GB extra memory, so the chosen value is a manual efficiency trade-off.
dconv = 4
Convolution kernel size in the Mamba blocks, set by hand following Mamba defaults in Section III-C.
expand = 2
Expansion factor for the Mamba state-space blocks, chosen without an ablation study in Section III-C.
num_blocks = 2
The number of BiMamba2DAC blocks is fixed at 2; no ablation is given for this architectural choice.
Frontend parameters (mel filters, STFT window, hop, sample rate) = 64 mel bands, 40ms window, 20ms hop, 24kHz
Audio front-end settings in Section III-C are chosen by hand and are not varied.
Training hyperparameters (learning rate, weight decay, batch size, epochs) = lr=1e-4, weight decay 1e-4 to 5e-6, batch 32-256, 120 epochs
Training schedule in Section III-C is set manually; performance can depend on these values, but they are not fitted to the reported metric.

assumptions (4)

domain assumption Equation (1) pseudo-FOA conversion preserves enough spatial information for models pretrained on real FOA signals to transfer to stereo input.
The method collapses stereo L/R into W/Y with X=Z=0, and no experiment validates that this preserves the directional cues needed for SELD. This is the main enabler of the transfer approach.
domain assumption Pretrained CNN14 and PSELDnet features generalize from their pretraining data to DCASE2025 stereo scenes.
The fine-tuning comparison assumes the pretrained representation is useful on the target domain; the paper gives no feature-distribution analysis.
domain assumption Checkpoint selection by highest validation location-dependent F1 is representative of true model quality on the evaluation set.
Section III-C selects one checkpoint per model; no repeated runs or confidence intervals are reported, and Table IV shows a large drop from development to official evaluation.
standard math The Mamba and Mamba2 discretization and selective state-space machinery work as described by the cited libraries.
The theory in Section II-A is taken from cited Mamba papers; the implementation is not independently verified in this work.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing Stereo Sound Event Detection with BiMamba and Pretrained PSELDnet." pith.science (2026). https://pith.science/paper/HUQ3TFZV

@misc{pith2026250709570,
  author       = {Pith},
  title        = {Pith review of: Enhancing Stereo Sound Event Detection with BiMamba and Pretrained PSELDnet},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HUQ3TFZV}},
  note         = {Machine review of arXiv:2507.09570}
}

read the original abstract

Pre-training methods have greatly improved the performance of sound event localization and detection (SELD). However, existing Transformer-based models still face high computational cost. To solve this problem, we present a stereo SELD system using a pre-trained PSELDnet and a bidirectional Mamba sequence model. Specifically, we replace the Conformer module with a BiMamba module. We also use asymmetric convolutions to better capture the time and frequency relationships in the audio signal. Test results on the DCASE2025 Task 3 development dataset show that our method performs better than both the baseline and the original PSELDnet with a Conformer decoder. In addition, the proposed model costs fewer computing resources than the baselines. These results show that the BiMamba architecture is effective for solving key challenges in SELD tasks. The source code is publicly accessible at https://github.com/ alexandergwm/DCASE2025 TASK3 Stereo PSELD Mamba.

Figures

Figures reproduced from arXiv: 2507.09570 by the authors.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

34 extracted references · 22 canonical work pages

[1]

Sound Event Localization and Detection of Overlap- ping Sources Using Convolutional Recurrent Neural Networks,

S. Adavanne, A. Politis, J. Nikunen, and T. Virtanen, “Sound Event Localization and Detection of Overlap- ping Sources Using Convolutional Recurrent Neural Networks,” arXiv, Jun. 2018. eprint: 1807.00129

arXiv 2018
[2]

Context-dependent sound event detection,

T. Heittola, A. Mesaros, A. Eronen, and T. Virtanen, “Context-dependent sound event detection,” J. AUDIO SPEECH MUSIC PROC. , vol. 2013, no. 1, pp. 1–13, Dec. 2013, ISSN : 1687-4722

work page 2013
[3]

Acoustic event detection in real life recordings,

A. Mesaros, T. Heittola, A. Eronen, and T. Virtanen, “Acoustic event detection in real life recordings,” in 2010 18th European Signal Processing Conference , 2010, pp. 1267–1271

work page 2010
[4]

FMSG- JLESS submission for DCASE 2024 task4 on sound event detection with heterogeneous training dataset and potentially missing labels,

Y . Xiao, H. Yin, J. Bai, and R. K. Das, “FMSG- JLESS submission for DCASE 2024 task4 on sound event detection with heterogeneous training dataset and potentially missing labels,” DCASE 2024 Challenge, Tech. Rep., 2024

work page 2024
[5]

WildDESED: An LLM- powered dataset for wild domestic environment sound event detection system,

Y . Xiao and R. K. Das, “WildDESED: An LLM- powered dataset for wild domestic environment sound event detection system,” in Proc. Workshop on Detec- tion and Classification of Acoustic Scenes and Events (DCASE), 2024, pp. 196–200

work page 2024
[6]

UCIL: An Unsupervised Class Incremental Learning Approach for Sound Event Detection,

Y . Xiao and R. K. Das, “UCIL: An Unsupervised Class Incremental Learning Approach for Sound Event Detection,” arXiv:2407.03657, 2024

arXiv 2024
[7]

Mixstyle based domain generalization for sound event detec- tion with heterogeneous training data,

Y . Xiao, H. Yin, J. Bai, and R. K. Das, “Mixstyle based domain generalization for sound event detec- tion with heterogeneous training data,” arXiv preprint arXiv:2407.03654, 2024

arXiv 2024
[8]

Multiple emitter location and signal pa- rameter estimation,

R. Schmidt, “Multiple emitter location and signal pa- rameter estimation,” IEEE Trans. Antennas Propag. , vol. 34, no. 3, pp. 276–280, Mar. 1986

work page 1986

Show all 34 references

[9]

Xiao and R

Y . Xiao and R. K. Das, Where’s that voice coming? continual learning for sound source localization , 2025. arXiv: 2407.03661 [eess.AS]. [Online]. Available: https://arxiv.org/abs/2407.03661

2025 arXiv
[10]

An Improved Event-Independent Net- work for Polyphonic Sound Event Localization and Detection,

Y . Cao, T. Iqbal, Q. Kong, F. An, W. Wang, and M. D. Plumbley, “An Improved Event-Independent Net- work for Polyphonic Sound Event Localization and Detection,” in ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, 2020,...

2021
[11]

ACCDOA: Activity-Coupled Carte- sian Direction of Arrival Representation for Sound Event Localization and Detection,

K. Shimada, Y . Koyama, N. Takahashi, S. Takahashi, and Y . Mitsufuji, “ACCDOA: Activity-Coupled Carte- sian Direction of Arrival Representation for Sound Event Localization and Detection,” arXiv, Oct. 2020. eprint: 2010.15306

2020 arXiv
[12]

Multi-ACCDOA: Local- izing and Detecting Overlapping Sounds from the Same Class with Auxiliary Duplicating Permutation Invariant Training,

K. Shimada, Y . Koyama, S. Takahashi, N. Takahashi, E. Tsunoo, and Y . Mitsufuji, “Multi-ACCDOA: Local- izing and Detecting Overlapping Sounds from the Same Class with Auxiliary Duplicating Permutation Invariant Training,” arXiv, Oct. 2021. eprint: 2110.07124

2021 arXiv
[13]

A Four-Stage Data Augmentation Approach to ResNet- Conformer Based Acoustic Modeling for Sound Event Localization and Detection,

Q. Wang, J. Du, H.-X. Wu, J. Pan, F. Ma, and C.-H. Lee, “A Four-Stage Data Augmentation Approach to ResNet- Conformer Based Acoustic Modeling for Sound Event Localization and Detection,” arXiv, Jan. 2021. eprint: 2101.02919

2021 arXiv
[14]

PSELDNets: Pre-trained Neural Networks on Large-scale Synthetic Datasets for Sound Event Localization and Detection,

J. Hu, Y . Cao, M. Wu, et al., “PSELDNets: Pre-trained Neural Networks on Large-scale Synthetic Datasets for Sound Event Localization and Detection,” arXiv, Nov
[15]

PANNs: Large-Scale Pretrained Au- dio Neural Networks for Audio Pattern Recognition,

Q. Kong, Y . Cao, T. Iqbal, Y . Wang, W. Wang, and M. D. Plumbley, “PANNs: Large-Scale Pretrained Au- dio Neural Networks for Audio Pattern Recognition,” arXiv, Dec. 2019. eprint: 1912.10211

2019 arXiv
[16]

Efficient Training of Audio Transformers with Patchout,

K. Koutini, J. Schl ¨uter, H. Eghbal-zadeh, and G. Wid- mer, “Efficient Training of Audio Transformers with Patchout,” arXiv, Oct. 2021. eprint: 2110.05069

2021 arXiv
[17]

HTS-AT: A Hierarchical Token- Semantic Audio Transformer for Sound Classification and Detection,

K. Chen, X. Du, B. Zhu, Z. Ma, T. Berg-Kirkpatrick, and S. Dubnov, “HTS-AT: A Hierarchical Token- Semantic Audio Transformer for Sound Classification and Detection,” arXiv, Feb. 2022. eprint: 2202.00874

2022 arXiv
[18]

Mamba: Linear-Time Sequence Modeling with Selective State Spaces,

A. Gu and T. Dao, “Mamba: Linear-Time Sequence Modeling with Selective State Spaces,” arXiv, Dec
[19]

Transformers are SSMs: General- ized Models and Efficient Algorithms Through Struc- tured State Space Duality,

T. Dao and A. Gu, “Transformers are SSMs: General- ized Models and Efficient Algorithms Through Struc- tured State Space Duality,” arXiv, May 2024. eprint: 2405.21060

2024 arXiv
[20]

SPMamba: State- space model is all you need in speech separation,

K. Li, G. Chen, R. Yang, and X. Hu, “SPMamba: State- space model is all you need in speech separation,”arXiv, Apr. 2024. eprint: 2404.02063

2024 arXiv
[21]

Xiao and R

Y . Xiao and R. K. Das, Tf-mamba: A time-frequency network for sound source localization , 2025. arXiv: 2409 . 05034 [eess.AS]. [Online]. Available: https : //arxiv.org/abs/2409.05034

2025 arXiv
[22]

XLSR-Mamba: a dual-column bidirectional state space model for spoofing attack detection,

Y . Xiao and R. K. Das, “XLSR-Mamba: a dual-column bidirectional state space model for spoofing attack detection,” IEEE Signal Processing Letters , vol. 32, pp. 1276–1280, 2025

2025
[23]

Mamba in Speech: Towards an Alternative to Self-Attention,

X. Zhang, Q. Zhang, H. Liu, et al., “Mamba in Speech: Towards an Alternative to Self-Attention,” arXiv, May
[24]

SELD-Mamba: Selective State-Space Model for Sound Event Localization and Detection with Source Distance Estimation,

D. Mu, Z. Zhang, H. Yue, Z. Wang, J. Tang, and J. Yin, “SELD-Mamba: Selective State-Space Model for Sound Event Localization and Detection with Source Distance Estimation,” arXiv, Aug. 2024. eprint: 2408.05057

2024 arXiv
[25]

Fonseca, X

E. Fonseca, X. Favory, J. Pons, F. Font, and X. Serra, “FSD50K,” Zenodo, 2025

2025
[26]

Image method for efficiently simulating small-room acoustics,

J. B. Allen and D. A. Berkley, “Image method for efficiently simulating small-room acoustics,” J. Acoust. Soc. Am., vol. 65, no. 4, pp. 943–950, Apr. 1979, ISSN : 0001-4966

1979
[27]

Rafaely, Fundamentals of Spherical Array Process- ing

B. Rafaely, Fundamentals of Spherical Array Process- ing. Cham, Switzerland: Springer International Publish- ing, 2019, ISBN : 978-3-319-99561-8. [Online]. Avail- able: https://link.springer.com/book/10.1007/978- 3- 319-99561-8. 5

2019 doi
[28]

Efficiently Modeling Long Sequences with Structured State Spaces,

A. Gu, K. Goel, and C. R ´e, “Efficiently Modeling Long Sequences with Structured State Spaces,” arXiv, Oct
[29]

DCASE2025 Task3 Stereo SELD Dataset,

K. Shimada, A. Politis, I. R. Roman, et al. , “DCASE2025 Task3 Stereo SELD Dataset,” Zenodo, Jun. 2025

2025
[30]

STARSS23: Sony-TAu Realistic Spatial Soundscapes 2023,

A. Politis, K. Shimada, P. Sudarsanam, et al. , “STARSS23: Sony-TAu Realistic Spatial Soundscapes 2023,” Zenodo, Mar. 2023

2023
[31]

A multi-level feature extraction network for sound event localization and detection in dcase 2025 task 3,

Q. Wan, Y . Hu, J. Liu, et al. , “A multi-level feature extraction network for sound event localization and detection in dcase 2025 task 3,” DCASE2025 Challenge, Tech. Rep., Jun. 2025

2025
[32]

Enhancing stereo sound event localization and detection through pre- trained audio representations and hybrid architectures,

T. Zhao, Z. Han, and M. Liu, “Enhancing stereo sound event localization and detection through pre- trained audio representations and hybrid architectures,” DCASE2025 Challenge, Tech. Rep., Jun. 2025

2025
[33]

Resnet-conformer for stereo sound event localization and distance estimation in dcase 2025 task3,

J. Park, H. Nam, and Y .-H. Park, “Resnet-conformer for stereo sound event localization and distance estimation in dcase 2025 task3,” DCASE2025 Challenge, Tech. Rep., Jun. 2025

2025
[34]

A conformer-based ensem- ble approach for sound event localization and detection for stereo data,

A. Bahuguna and R. Peter, “A conformer-based ensem- ble approach for sound event localization and detection for stereo data,” DCASE2025 Challenge, Tech. Rep., Jun. 2025. 6

2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Sound Event Localization and Detection of Overlap- ping Sources Using Convolutional Recurrent Neural Networks,

S. Adavanne, A. Politis, J. Nikunen, and T. Virtanen, “Sound Event Localization and Detection of Overlap- ping Sources Using Convolutional Recurrent Neural Networks,” arXiv, Jun. 2018. eprint: 1807.00129

arXiv 2018

[2] [2]

Context-dependent sound event detection,

T. Heittola, A. Mesaros, A. Eronen, and T. Virtanen, “Context-dependent sound event detection,” J. AUDIO SPEECH MUSIC PROC. , vol. 2013, no. 1, pp. 1–13, Dec. 2013, ISSN : 1687-4722

work page 2013

[3] [3]

Acoustic event detection in real life recordings,

A. Mesaros, T. Heittola, A. Eronen, and T. Virtanen, “Acoustic event detection in real life recordings,” in 2010 18th European Signal Processing Conference , 2010, pp. 1267–1271

work page 2010

[4] [4]

FMSG- JLESS submission for DCASE 2024 task4 on sound event detection with heterogeneous training dataset and potentially missing labels,

Y . Xiao, H. Yin, J. Bai, and R. K. Das, “FMSG- JLESS submission for DCASE 2024 task4 on sound event detection with heterogeneous training dataset and potentially missing labels,” DCASE 2024 Challenge, Tech. Rep., 2024

work page 2024

[5] [5]

WildDESED: An LLM- powered dataset for wild domestic environment sound event detection system,

Y . Xiao and R. K. Das, “WildDESED: An LLM- powered dataset for wild domestic environment sound event detection system,” in Proc. Workshop on Detec- tion and Classification of Acoustic Scenes and Events (DCASE), 2024, pp. 196–200

work page 2024

[6] [6]

UCIL: An Unsupervised Class Incremental Learning Approach for Sound Event Detection,

Y . Xiao and R. K. Das, “UCIL: An Unsupervised Class Incremental Learning Approach for Sound Event Detection,” arXiv:2407.03657, 2024

arXiv 2024

[7] [7]

Mixstyle based domain generalization for sound event detec- tion with heterogeneous training data,

Y . Xiao, H. Yin, J. Bai, and R. K. Das, “Mixstyle based domain generalization for sound event detec- tion with heterogeneous training data,” arXiv preprint arXiv:2407.03654, 2024

arXiv 2024

[8] [8]

Multiple emitter location and signal pa- rameter estimation,

R. Schmidt, “Multiple emitter location and signal pa- rameter estimation,” IEEE Trans. Antennas Propag. , vol. 34, no. 3, pp. 276–280, Mar. 1986

work page 1986