Pith. sign in

REVIEW 4 major objections 4 minor 49 references

Self-Steering Deep Non-Linear Spatially Selective Filters for Efficient Extraction of Moving Speakers under Weak Guidance

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Feedback loop between tracker and enhancer cuts moving-speaker tracking error from 9.8 to 2.9 degrees.

desk verdict A plausible closed-loop tracker/SSF design with a real but addressable self-confirmation concern; add the MIMO-concat ablation and a perturbation test and it's a solid TSE paper. read the letter →

arxiv 2507.02791 v2 pith:ERZAIDC3 submitted 2025-07-03 eess.AS cs.LGcs.SD

Jakob Kienegger , Alina Mannanova , Huajian Fang , Timo Gerkmann This is my paper · ORCID
classification eess.AScs.LGcs.SD
keywords movingspeakerextractionweaklyguidedtargetparticlefilterautoregressivefeedbackspatiallyselectiveMIMOmaskreal-timespeechenhancementDoAtracking
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 claims that a moving speaker can be extracted from noisy multi-microphone mixtures using only the speaker's starting direction as guidance, by closing a loop between a lightweight particle filter and a deep non-linear spatially selective filter. The key move is to compute a multi-channel mask that preserves the target's spatial propagation, then feed the filter's enhanced output back to the tracker as its observation. In a synthetic two-speaker test, median angular error drops from 9.78 degrees to 2.88 degrees, median SI-SDR rises from 3.05 dB to 3.52 dB, and PESQ from 1.50 to 1.61, relative to the concatenative baseline. A listening test on recorded movement scenarios shows a preference for the self-steering pipeline over both compared frameworks.

What carries the argument

The central mechanism is the autoregressive integration of a bootstrap particle filter with the FT-JNF spatially selective filter. The SSF is modified to output a MIMO mask $\mathbf{M}_{tk}$ so that the enhanced signal $\hat{\mathbf{S}}_{tk} = \mathbf{M}_{tk} \odot \mathbf{Y}_{tk}$ preserves spatial cues; the particle filter then evaluates its likelihood as the output power of a delay-and-sum beamformer applied to this enhanced signal, using only the previous frame's DoA estimate to steer the mask (a one-frame lookahead). A short end-to-end fine-tuning with mean angular error as an auxiliary loss aligns the filter's output with the tracker's needs.

What would settle it

Run the pipeline while deliberately steering the SSF with a fixed, wrong direction (for instance, the true DoA plus 30 degrees) and check whether the particle filter's posterior drifts toward the true speaker or toward the wrong steering direction; if it drifts toward the steering direction, the tracking improvement is at least partly self-confirmation rather than true localization.

Watch

Extended reading notes

Core claim

The paper establishes that tracking and enhancement need not be separate stages in weak-guidance target speaker extraction. When the spatially selective filter is adapted from a single-reference-channel mask to a multi-channel mask, its output retains the target's direct-path inter-microphone structure; that output, when used by a bootstrap particle filter through a delay-and-sum beamformer likelihood, becomes a much more informative observation than the raw microphone mixture. Because the filter's mask is computed from the previous DoA estimate, the system is autoregressive, and the paper shows empirically that this loop converges to accurate DoA estimates and strong enhancement, with the particle filter's median angular error shrinking from 9.78 degrees to 2.88 degrees, and both SI-SDR and PESQ improving accordingly.

Load-bearing premise

The enhanced signal carries the true direction of the target even though the filter that produced it was steered using the tracker's previous estimate, rather than the current true direction.

Editorial extensions

If this is right

  • Real-time moving-speaker extraction no longer requires a data-driven tracker; a 50-particle bootstrap filter can do the job when fed the enhanced output.
  • The MIMO mask formulation doubles as a spatial-feature-preserving front-end, so localization and enhancement share one lightweight computation.
  • With only a handful of fine-tuning steps (25 epochs) the autoregressive loop trains stably, avoiding the non-parallelizable training of typical AR architectures.
  • The benefit is largest for far-field speakers under low SNR, where spatial information matters most.
  • Because the method leans more heavily on spatial cues than the concatenative baseline, its success depends on the beamformer-power likelihood retaining those cues in reverberant conditions.

Reading between the lines

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

  • The reported angular error may be partly inflated by self-confirmation: an SSF steered off-target could shape the enhanced signal's spatial statistics toward its own steering direction, rewarding the filter for persisting in its belief rather than finding the speaker.
  • The general recipe — a cheap Bayes filter whose observation is the output of a differentiable enhancement front-end, fine-tuned with an auxiliary localization loss — could transfer to other state estimation problems such as tracking multiple speakers or moving noise sources.
  • An oracle experiment that feeds the PF the true enhanced signal (from oracle DoAs) would quantify the ceiling of the autoregressive loop and separate tracking gains from enhancement gains.
  • Real-world deployment will need an adaptive motion model or a bank of models, since the constant-velocity model is matched to the synthetic training trajectories rather than learned from data.
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

4 major / 4 minor

Summary. The paper proposes a weakly guided target-speaker extraction pipeline for moving speakers. It replaces strong per-frame direction cues with knowledge of only the initial azimuth, using a particle filter (PF) to estimate the speaker's trajectory and steer a recurrent spatially selective filter (FT-JNF). Its novel ingredient is temporal feedback: the SSF computes a multi-channel mask using the PF's previous DoA estimate (Eq. 6), and the PF likelihood is evaluated on the SSF's enhanced output rather than on the raw mixture (Section 5.2). The authors evaluate MISO and MIMO SSF variants on a synthetic two-speaker dataset and a real-world recording listening test, reporting large median angular-error reductions (e.g., 9.78° to 2.88° for PF-CV in Table 1) and modest enhancement gains (SI-SDR 3.05 to 3.52 dB; PESQ 1.50 to 1.61). The central claim, stated in Section 7, is that the resulting self-steering pipeline drastically improves tracking accuracy and yields strong enhancement performance.

Significance. If the closed-loop formulation is genuinely responsible for the reported gains, the result is significant for resource-constrained real-time enhancement: it would show that a lightweight PF can achieve tracking accuracy close to oracle steering without a data-driven tracker. The paper has several strengths: an oracle comparison showing that MIMO does not outperform MISO (Table 1, rows 1-2), a mismatched motion-model condition (PF-RW), a real-world listening test, and a closed-form motion-model calibration (Eq. 8). However, the central evidence is weakened by the absence of a MIMO concatenative baseline and by the self-referential nature of the PF measurement model. The significance of the contribution is therefore contingent on additional control experiments that separate the effect of autoregressive feedback from the effect of the multi-channel mask and that probe the closed loop under incorrect initialization.

major comments (4)
  1. [Section 4, Eqs. (5)-(6); Section 5.2] The PF observation is not independent of the tracker's own state: the SSF mask M_t is computed from (Y_t, θ_{t-1}), and the likelihood is then computed from a delay-and-sum beamformer applied to S_t = M_t ⊙ Y_t. Because the mask attenuates directions away from θ_{t-1}, the beamformer power will tend to peak near the previous estimate regardless of the true source direction. This violates the standard state-space assumption behind Eq. (4) that observations depend on the current state x_t, and it creates a self-confirming loop. The tracking gains in Table 1 may therefore reflect lock-on to the filter's own steering direction rather than genuine localization. The manuscript contains no experiment that probes the loop under a wrong initialization or a temporarily lost track; I would like to see an initialization-offset test (e.g., starting the PF at a deliberately wrong angle and measuring the transient recovery behavior) and an analysis of tracking failures rather than only averaged angular error.
  2. [Table 1] The comparison supporting the central claim is confounded. Rows (3) and (5) are MISO concatenative systems, while rows (4) and (6) are MIMO autoregressive systems. Without a MIMO PF-CV concatenative baseline, the reader cannot attribute the improvement to the temporal feedback rather than to the multi-channel mask formulation. The oracle comparison (rows 1-2) mitigates this concern by showing that MIMO alone is not better than MISO, but it does not resolve the confound because the weakly guided MIMO system is never evaluated in concatenative mode. Adding a MIMO PF-CV concatenative condition is a necessary control for the paper's claim that the autoregressive interplay drives the improvement.
  3. [Section 6, Fig. 4b] The listening-test conclusion is presented as a 'clear tendency' and the abstract and conclusion describe 'strong enhancement performance,' but the subjective evidence is based on 10 participants and 12 test cases with no significance test or inter-rater agreement. In addition, the objective enhancement gains in Table 1 are modest (about 0.5 dB in median SI-SDR and about 0.1 in median PESQ). The subjective evidence should be described as suggestive rather than demonstrative, and the strength claims in the abstract and Section 7 should be tempered accordingly unless significance testing or confidence intervals are provided.
  4. [Section 5.2] The PF hyperparameters (motion variance, beamformer power weighting, observation noise, and the number of particles N=50) are said to be determined by an exhaustive search on the validation subset, but the exact values are not reported. The weight of the MAE auxiliary loss used during fine-tuning is also not reported. This limits reproducibility and makes the claimed low computational overhead hard to verify independently.
minor comments (4)
  1. [Section 6, first paragraph] There is a typo: 'more then 40°' should read 'more than 40°'.
  2. [Section 5.1, Eq. (8)] The sentence 'Due to the steady increase of the expected value in (8), we will only use trajectories of 5 s length' is a modeling choice that is not further justified; a brief explanation of how trajectory length interacts with the motion-model calibration would improve clarity.
  3. [Section 3.1 and Eq. (5)] The change from the MISO extraction in Eq. (2) to the MIMO extraction in Eq. (5) changes the quantity being estimated from a single-channel reference-microphone signal to a multi-channel direct-path signal. The paper should state explicitly whether the evaluation metrics are computed on the same reference channel for both formulations, since oracle MIMO gives lower SI-SDR than oracle MISO.
  4. [Section 5.2] The statement that the MAE auxiliary loss 'consistently improved convergence' is not supported by a quantitative comparison or ablation; reporting the validation performance with and without this loss would make the claim verifiable.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the AR feedback is explicit and the main results are evaluated against ground truth.

full rationale

The paper's central claim is that feeding the SSF output back to a PF improves tracking and enhancement. This is not a circular derivation: Eq. (6) explicitly defines the SSF mask as a function of the previous PF estimate and the current mixture, and the PF likelihood is then computed on the SSF output via beamformer power (Sec. 5.2). The dependence of the observation on its own previous estimate is a deliberate closed-loop design, not a hidden equivalence; the paper does not define the PF's success metric in terms of the feedback. Tracking accuracy is measured as angular error against ground-truth trajectories (Table 1), and the comparison to concatenative PF baselines (PF-RW concat, PF-CV concat) provides an external control. The self-citations to [9] supply dataset construction, motion-model choices, and prior fine-tuning findings; they are motivational and methodological, not load-bearing uniqueness claims or ansatze invoked to preclude alternatives. A potential self-confirmation bias in the closed-loop observation model is a plausible empirical concern (e.g., the mask may pass energy near the previous DoA), and an initialization-perturbation test would strengthen the paper, but this is a robustness/specification issue, not a definitional reduction of the predicted quantity to the fitted input. No fitted parameter is renamed as a prediction, and no prior result by the same authors is used as the sole justification for the main conclusion.

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

No new physical entities, forces, or conserved quantities are postulated. The free parameters are standard system-design choices such as particle count, motion model noise, likelihood scaling, and fine-tuning loss weights. The key domain assumption is that the enhanced SSF output preserves enough spatial information to serve as a valid observation for the particle filter; this assumption is load-bearing and largely unmeasured.

free parameters (5)
  • Particle filter motion model parameters (CV sigma, RW diffusion) = not disclosed, chosen by exhaustive search on validation subset
    Section 5.2: 'we fix the number of particles N to 50 and optimize the remaining model parameters with an exhaustive search on the validation subset.' These parameters directly affect tracking accuracy and therefore the reported gains.
  • Particle filter likelihood parameters (beamformer power weighting, observation noise variance) = not disclosed, same exhaustive search
    The likelihood is constructed from the output power of a delay-and-sum beamformer [18]; its scaling or noise parameters are not reported, which affects how strongly observations influence the tracker.
  • MAE auxiliary loss weight during fine-tuning = not reported
    Section 5.2 mentions including mean angular error as an auxiliary loss; the weighting is not given, and it shapes the SSF output that is later used as the PF observation.
  • Number of particles N = 50
    Section 5.2: 'We fix the number of particles N to 50.' This is a hand-set design choice that trades off computational cost against tracking accuracy.
  • Sigma of azimuthal motion perturbation in Eq. (7) = set so that expected absolute velocity is 1.5 m/s at trajectory end
    Section 5.1, Eq. (8): the closed-form relation is used to calibrate sigma to a medium walking speed [24, 25]. This is a modeling choice rather than a value fitted to enhancement performance.
assumptions (5)
  • domain assumption The relative azimuth theta_t is a sufficient cue to separate the target speaker from interfering speakers.
    Used throughout Section 3.1 and Eqs. (2)-(3); the SSF is steered only by DoA information, with no speaker identity embedding.
  • domain assumption The MIMO SSF output retains the target's direct-path inter-microphone propagation structure, so a beamformer power likelihood on that output is informative about the true DoA.
    Section 4, Eqs. (5)-(6): this is the mechanism that justifies feeding the SSF output to the PF; the paper does not measure how accurately the enhanced signal preserves spatial cues.
  • standard math Bootstrap particle filtering with 50 particles and a beamformer power likelihood approximates the posterior p(x_t | S_1:t, x_0) well enough for the SSF.
    Section 3.2, Eq. (4): standard sequential Monte Carlo approximation [20-23], but the particle count and proposal are design choices that may limit accuracy.
  • domain assumption The synthetic CV trajectory model and its speed calibration represent natural human motion, and fine-tuning on this synthetic distribution transfers to real rooms.
    Section 5.1, Eqs. (7)-(8), and Section 5.2: evaluation uses a matched CV model and a mismatched RW model, but the real recordings are not objectively labeled for DoA.
  • ad hoc to paper Fine-tuning with an MAE auxiliary loss reaches a useful fixed point rather than collapsing the SSF into echoing the PF's previous direction.
    Section 5.2: the AR loop is trained with a small number of fine-tuning steps 'yielding the best performance', but no stability analysis of the feedback loop is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Self-Steering Deep Non-Linear Spatially Selective Filters for Efficient Extraction of Moving Speakers under Weak Guidance." pith.science (2026). https://pith.science/paper/ERZAIDC3

@misc{pith2026250702791,
  author       = {Pith},
  title        = {Pith review of: Self-Steering Deep Non-Linear Spatially Selective Filters for Efficient Extraction of Moving Speakers under Weak Guidance},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ERZAIDC3}},
  note         = {Machine review of arXiv:2507.02791}
}
read the original abstract

Recent works on deep non-linear spatially selective filters demonstrate exceptional enhancement performance with computationally lightweight architectures for stationary speakers of known directions. However, to maintain this performance in dynamic scenarios, resource-intensive data-driven tracking algorithms become necessary to provide precise spatial guidance conditioned on the initial direction of a target speaker. As this additional computational overhead hinders application in resource-constrained scenarios such as real-time speech enhancement, we present a novel strategy utilizing a low-complexity tracking algorithm in the form of a particle filter instead. Assuming a causal, sequential processing style, we introduce temporal feedback to leverage the enhanced speech signal of the spatially selective filter to compensate for the limited modeling capabilities of the particle filter. Evaluation on a synthetic dataset illustrates how the autoregressive interplay between both algorithms drastically improves tracking accuracy and leads to strong enhancement performance. A listening test with real-world recordings complements these findings by indicating a clear trend towards our proposed self-steering pipeline as preferred choice over comparable methods.

Figures

Figures reproduced from arXiv: 2507.02791 by the authors.

Figure 1
Figure 1. Weakly guided target speaker extraction ( [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Synthetic two-speaker ( / ) dataset with continuous movement. The motion model reflects natural human walking speed [24], [25]. 5. EXPERIMENTS 5.1. Dataset Synthetic dataset We employ a synthetic dataset consisting of noisy and reverberant two-speaker mixtures to enable NN training, model optimization and evaluation in a controlled acoustic scenario. In particular, we use utterances from the LibriSpeech corpus [29] … view at source ↗
Figure 4
Figure 4. Objective (a) and subjective (b) evaluation of speech enhance [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (1 more)
Figure 3
Figure 3. Figure 3: Dependency of enhancement performance (SI-SDR) on tracking accuracy (MAE). Solid lines display a log-linear regression model. 2 6 10 14 18 ∆SI-SDR [dB] → PF-RW PF-CV 0.0 0.5 1.0 1.5 ∆PESQ → concat. AR (ours) (a) Intrusive metrics representing speech quality (PESQ) and …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 48 canonical work pages

  1. [1]

    Neural target speech extraction: An overview,

    K. Zmolikova, M. Delcroix, T. Ochiai, K. Kinoshita, J. ˇCernock´y, and D. Yu, “Neural target speech extraction: An overview,” IEEE Signal Proc. Magazine, vol. 40, 2023

    work page 2023

  2. [2]

    Multi-channel speech separation using spatially selective deep non-linear filters,

    K. Tesch and T. Gerkmann, “Multi-channel speech separation using spatially selective deep non-linear filters,” IEEE/ACM TASLP, vol. 32, 2024

    work page 2024

  3. [3]

    Exploiting spatial information with the informed complex-valued spatial autoencoder for target speaker extraction,

    A. Briegleb, M. M. Halimeh, and W. Kellermann, “Exploiting spatial information with the informed complex-valued spatial autoencoder for target speaker extraction,” in IEEE ICASSP , 2023

    work page 2023

  4. [4]

    Spatially selective speaker separation using a DNN with a location dependent feature extraction,

    A. Bohlender, A. Spriet, W. Tirry, and N. Madhu, “Spatially selective speaker separation using a DNN with a location dependent feature extraction,” IEEE/ACM TASLP, vol. 32, 2024

    work page 2024

  5. [5]

    All neural low-latency directional speech extraction,

    A. Pandey, S. Lee, J. Azcarreta, D. Wong, and B. Xu, “All neural low-latency directional speech extraction,” in Interspeech, 2024

    work page 2024

  6. [6]

    ReZero: Region-customizable sound extraction,

    R. Gu and Y . Luo, “ReZero: Region-customizable sound extraction,” IEEE/ACM TASLP, 2024

    work page 2024

  7. [7]

    Continuous speech separation: Dataset and analysis,

    Z. Chen, T. Yoshioka, L. Lu, T. Zhou, Z. Meng, Y . Luo, J. Wu, X. Xiao, and J. Li, “Continuous speech separation: Dataset and analysis,” in IEEE ICASSP, 2020

    work page 2020

  8. [8]

    The fifth ’CHiME’ speech separation and recognition challenge: Dataset, task and baselines,

    J. Barker, S. Watanabe, E. Vincent, and J. Trmal, “The fifth ’CHiME’ speech separation and recognition challenge: Dataset, task and baselines,” in Interspeech, 2018

    work page 2018

Show all 49 references
  1. [9]

    Steering deep non-linear spatially selective filters for weakly guided extraction of moving speakers in dynamic scenarios,

    J. Kienegger and T. Gerkmann, “Steering deep non-linear spatially selective filters for weakly guided extraction of moving speakers in dynamic scenarios,” in Interspeech, 2025

    2025

  2. [10]

    A convolutional recurrent neural network for real-time speech enhancement,

    K. Tan and D. Wang, “A convolutional recurrent neural network for real-time speech enhancement,” in Interspeech, 2018

    2018

  3. [11]

    Real time speech enhancement in the waveform domain,

    A. D ´efossez, G. Synnaeve, and Y . Adi, “Real time speech enhancement in the waveform domain,” in Interspeech, 2020

    2020

  4. [12]

    Towards efficient models for real-time deep noise suppression,

    S. Braun, H. Gamper, C. K. Reddy, and I. Tashev, “Towards efficient models for real-time deep noise suppression,” in IEEE ICASSP , 2021

    2021

  5. [13]

    Conv-TasNet: Surpassing ideal time–frequency magnitude masking for speech separation,

    Y . Luo and N. Mesgarani, “Conv-TasNet: Surpassing ideal time–frequency magnitude masking for speech separation,” IEEE/ACM Transactions on Audio, Speech, and Language Processing , vol. 27, 2019

    2019

  6. [14]

    An investigation of incorporating Mamba for speech enhancement,

    R. Chao, W.-H. Cheng, M. L. Quatra, S. M. Siniscalchi, C.-H. H. Yang, S.-W. Fu, and Y . Tsao, “An investigation of incorporating Mamba for speech enhancement,” in IEEE Spoken Language Tech. Workshop , 2024

    2024

  7. [15]

    Iterative autoregression: A novel trick to improve your low-latency speech enhancement model,

    P. Andreev, N. Babaev, A. Saginbaev, I. Shchekotov, and A. Alanov, “Iterative autoregression: A novel trick to improve your low-latency speech enhancement model,” in Interspeech, 2023

    2023

  8. [16]

    PARIS: Pseudo-autoregressive siamese training for online speech separation,

    Z. Pan, G. Wichern, F. G. Germain, K. Saijo, and J. Le Roux, “PARIS: Pseudo-autoregressive siamese training for online speech separation,” in Interspeech, 2024

    2024

  9. [17]

    Nonlinear filtering for speaker tracking in noisy and reverberant environments,

    J. Vermaak and A. Blake, “Nonlinear filtering for speaker tracking in noisy and reverberant environments,” in IEEE ICASSP , 2001

    2001

  10. [18]

    Particle filtering algorithms for tracking an acoustic source in a reverberant environment,

    D. Ward, E. Lehmann, and R. Williamson, “Particle filtering algorithms for tracking an acoustic source in a reverberant environment,” IEEE Trans. on Speech and Audio Proc. , vol. 11, 2003

    2003

  11. [19]

    Particle filter algorithm for DoA tracking using co-prime array,

    F. Dong, L. Xu, and X. Li, “Particle filter algorithm for DoA tracking using co-prime array,” IEEE Comm. Letters , vol. 24, 2020

    2020

  12. [20]

    Hagiwara, Time series analysis for the state-space model with R/Stan

    J. Hagiwara, Time series analysis for the state-space model with R/Stan . Springer, 2021

    2021

  13. [21]

    A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking,

    M. Arulampalam, S. Maskell, N. Gordon, and T. Clapp, “A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking,” IEEE Trans. on Signal Proc. , vol. 50, 2002

    2002

  14. [22]

    Real-time speaker tracking using particle filter sensor fusion,

    Y . Chen and Y . Rui, “Real-time speaker tracking using particle filter sensor fusion,” Proc. of the IEEE , vol. 92, 2004

    2004

  15. [23]

    Novel approach to nonlinear/non- Gaussian Bayesian state estimation,

    N. Gordon, D. Salmond, and A. Smith, “Novel approach to nonlinear/non- Gaussian Bayesian state estimation,” IEE proc. F , vol. 140, 1993

    1993

  16. [24]

    Relationship between vertical ground reaction force and speed during walking, slow jogging, and running,

    T. S. Keller, A. Weisberger, J. L. Ray, S. Hasan, R. G. Shiavi, and D. M. Spengler, “Relationship between vertical ground reaction force and speed during walking, slow jogging, and running,” Clinical biomechanics, vol. 11, 1996

    1996

  17. [25]

    Outdoor walking speeds of apparently healthy adults: A systematic review and meta-analysis,

    E. M. Murtagh, J. L. Mair, E. Aguiar, C. Tudor-Locke, and M. H. Murphy, “Outdoor walking speeds of apparently healthy adults: A systematic review and meta-analysis,” Sports Medicine , vol. 51, 2021

    2021

  18. [26]

    GCC-Speaker: Target speaker localization with optimal speaker-dependent weighting in multi-speaker scenarios,

    G. Li, W. Xue, W. Liu, J. Yi, and J. Tao, “GCC-Speaker: Target speaker localization with optimal speaker-dependent weighting in multi-speaker scenarios,” in IEEE ICASSP , 2023

    2023

  19. [27]

    LocSelect: Target speaker localization with an auditory selective hearing mechanism,

    Y . Chen, X. Qian, Z. Pan, K. Chen, and H. Li, “LocSelect: Target speaker localization with an auditory selective hearing mechanism,” in IEEE ICASSP , 2024

    2024

  20. [28]

    Robust frame-level speaker localization in reverberant and noisy en- vironments by exploiting phase difference losses,

    S. S. Battula, H. Taherian, A. Pandey, D. Wong, B. Xu, and D. Wang, “Robust frame-level speaker localization in reverberant and noisy en- vironments by exploiting phase difference losses,” in IEEE ICASSP , 2025

    2025

  21. [29]

    Librispeech: An ASR corpus based on public domain audio books,

    V . Panayotov, G. Chen, D. Povey, and S. Khudanpur, “Librispeech: An ASR corpus based on public domain audio books,” in IEEE ICASSP , 2015

    2015

  22. [30]

    Librimix: An open-source dataset for generalizable speech separation,

    J. Cosentino, M. Pariente, S. Cornell, A. Deleforge, and E. Vincent, “Librimix: An open-source dataset for generalizable speech separation,”

  23. [31]

    gpuRIR: A Python library for room impulse response simulation with GPU acceleration,

    D. Diaz-Guerra, A. Miguel, and J. R. Beltr ´an, “gpuRIR: A Python library for room impulse response simulation with GPU acceleration,” Multimedia Tools and Applications , vol. 80, 2018

    2018

  24. [32]

    Image method for efficiently simulating small- room acoustics,

    J. Allen and D. Berkley, “Image method for efficiently simulating small- room acoustics,” J. Acoust. Soc. Am. , vol. 65, 1979

    1979

  25. [33]

    On the role of spatial, spectral, and temporal processing for DNN-based non-linear multi- channel speech enhancement,

    K. Tesch, N.-H. Mohrmann, and T. Gerkmann, “On the role of spatial, spectral, and temporal processing for DNN-based non-linear multi- channel speech enhancement,” in Interspeech, 2022

    2022

  26. [34]

    Insights into deep non-linear filters for improved multi-channel speech enhancement,

    K. Tesch and T. Gerkmann, “Insights into deep non-linear filters for improved multi-channel speech enhancement,” IEEE/ACM TASLP, vol. 31, 2023

    2023

  27. [35]

    Spatially selective deep non-linear filters for speaker extraction,

    ——, “Spatially selective deep non-linear filters for speaker extraction,” in IEEE ICASSP , 2023

    2023

  28. [36]

    Survey of maneuvering target tracking. Part I. Dynamic models,

    X. Rong Li and V . Jilkov, “Survey of maneuvering target tracking. Part I. Dynamic models,” IEEE Trans. on Aerospace and Electronic Systems , vol. 39, 2003

    2003

  29. [37]

    Accurate adaptive filtering in square-root Hann windowed short-time fourier transform domain,

    S. Shimauchi and H. Ohmuro, “Accurate adaptive filtering in square-root Hann windowed short-time fourier transform domain,” in IEEE ICASSP , 2014

    2014

  30. [38]

    Generating sensor signals in isotropic noise fields,

    E. Habets and S. Gannot, “Generating sensor signals in isotropic noise fields,” J. Acoust. Soc. Am. , vol. 122, 2007

    2007

  31. [39]

    Fairbanks, V oice and Articulation Drillbook

    G. Fairbanks, V oice and Articulation Drillbook . Harper, 1960

    1960

  32. [40]

    Neural directional filtering: Far-field directivity control with a small microphone array,

    J. Wechsler, S. R. Chetupalli, M. M. Halimeh, O. Thiergart, and E. A. P. Habets, “Neural directional filtering: Far-field directivity control with a small microphone array,” in IWAENC, 2024

    2024

  33. [41]

    Spatially constrained vs. unconstrained filtering in neural spatiospectral filters for multichannel speech enhance- ment,

    A. Briegleb and W. Kellermann, “Spatially constrained vs. unconstrained filtering in neural spatiospectral filters for multichannel speech enhance- ment,” in EUSIPCO, 2024

    2024

  34. [42]

    Utilizing head rotation data in DNN-based multi-channel speech enhancement for hearing aids,

    B. Lentz and R. Martin, “Utilizing head rotation data in DNN-based multi-channel speech enhancement for hearing aids,” in IWAENC, 2024

    2024

  35. [43]

    Particle filter with integrated voice activity detection for acoustic source tracking,

    E. A. Lehmann and A. M. Johansson, “Particle filter with integrated voice activity detection for acoustic source tracking,” EURASIP J. on Adv. in Signal Proc. , vol. 2007, 2006

    2007

  36. [44]

    Doucet, N

    A. Doucet, N. De Freitas, N. J. Gordon et al. , Sequential Monte Carlo methods in practice . Springer, 2001

    2001

  37. [45]

    Particle filter networks with application to visual localization,

    P. Karkus, D. Hsu, and W. S. Lee, “Particle filter networks with application to visual localization,” in PMLR Conf. on Robot Learning , 2018

    2018

  38. [46]

    A wrapped Kalman filter for azimuthal speaker tracking,

    J. Traa and P. Smaragdis, “A wrapped Kalman filter for azimuthal speaker tracking,” IEEE Signal Proc. Letters , vol. 20, no. 12, 2013

    2013

  39. [47]

    Perceptual evaluation of speech quality (PESQ)-a new method for speech quality assessment of telephone networks and codecs,

    A. Rix, J. Beerends, M. Hollier, and A. Hekstra, “Perceptual evaluation of speech quality (PESQ)-a new method for speech quality assessment of telephone networks and codecs,” in IEEE ICASSP , 2001

    2001

  40. [48]

    SDR – Half- baked or well done?

    J. L. Roux, S. Wisdom, H. Erdogan, and J. R. Hershey, “SDR – Half- baked or well done?” in IEEE ICASSP , 2019

    2019

  41. [2020]

    Available: https://arxiv.org/abs/2005.11262

    [Online]. Available: https://arxiv.org/abs/2005.11262

    2005 arXiv

Pith tools

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