Pith. sign in

REVIEW 3 major objections 6 minor 36 references

End-to-End Multi-Microphone Speaker Extraction Using Relative Transfer Functions

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

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

Pith's one-line read An end-to-end multi-microphone extraction network conditioned on the instantaneous relative transfer function outperforms both oracle DOA and spectral enrollment cues.

desk verdict A modest but clean empirical comparison showing RTF enrollment beats oracle DOA and spectral cues for learned multichannel extraction; the same-DOA conclusion overreaches because the DOA baseline is missing. read the letter →

arxiv 2502.06285 v1 pith:5S6IACKB submitted 2025-02-10 cs.SD cs.AI

Aviad Eisenberg , Sharon Gannot , Shlomo E. Chazan This is my paper · ORCID
classification cs.SDcs.AI
keywords TargetspeakerextractionRelativetransferfunctionMulti-microphoneSpatialcueDirectionofarrivalDeepneuralnetworkseparationSI-SDR
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 attempts to establish that the instantaneous relative transfer function (RTF), computed from a noiseless reference utterance at the desired speaker's position, is a more effective spatial enrollment cue for end-to-end multi-microphone target speaker extraction than either oracle direction-of-arrival (DOA) or a single-channel spectral voice embedding. The authors compare these cues on simulated reverberant mixtures with four microphones and directional noise, and report that the RTF-conditioned network reaches 9.2 dB SI-SDR and 0.81 STOI, ahead of the DOA-conditioned variant at 8.4 dB and the spectral variant at 8.18 dB. They also show the RTF model still separates two speakers with the same DOA, which a DOA cue cannot do. If correct, the result indicates that a full spatial fingerprint of the source, including reverberation and distance information, can be injected directly into a learned separator without an explicit beamformer.

What carries the argument

The central object is the instantaneous relative transfer function, defined as the complex ratio of the STFT of the enrollment signal at each microphone to that at a reference microphone, $\hat{r}_d(t,k)=\tilde{s}_d^{\mathrm{ref}}(t,k)/\tilde{s}_d^{\mathrm{ref},m}(t,k)$. Because the enrollment is clean and recorded from the exact position of the desired source, this ratio acts as a non-blind estimate of the relative acoustic transfer function and encodes the full spatial response, including reverberation and multipath, in contrast to a single arrival angle. The network processes the mixture through a multi-channel encoder and the RTF through a separate encoder, averages the RTF embedding over the frame dimension, multiplies it frame-by-frame into the mixture embedding, and reconstructs the target through a decoder with six self-attention layers; training uses the SI-SDR loss with a swap-augmentation trick that averages the extraction loss for both speakers in each mixture.

What would settle it

Run the same training setup but corrupt the enrollment with additive noise or shift its position by a few centimeters; if TSE-RTF then falls behind TSE-DOA on the same test set, the claimed advantage depends on unrealistic enrollment conditions. Similarly, place the desired speaker on a short moving trajectory during the mixture; a static RTF should fail to track, exposing the time-invariance assumption.

Watch

Extended reading notes

Core claim

The paper's central claim is that a per-frequency instantaneous RTF, estimated as the ratio of the enrollment signal at each microphone to that at an arbitrary reference microphone, is a better conditioning signal for a multi-microphone extraction network than a one-hot or learned DOA embedding or a voice-signature spectral embedding. In their experiments, the proposed TSE-RTF model achieves 9.2 dB SI-SDR and 0.81 STOI on a random-location test set, outperforming TSE-DOA (8.4 dB, 0.80) and TSE-Spectral (8.18 dB, 0.80), and also outperforms the estimated MVDR beamformer and the single-channel baseline. In a same-DOA scenario, the RTF model keeps 8.8 dB SI-SDR, slightly above the oracle MVDR's 8.5 dB, though its STOI of 0.79 trails the oracle MVDR's 0.82. The authors interpret this as evidence that the instantaneous RTF encodes spatial information beyond the source angle, such as multipath and distance, and that this information survives end-to-end training.

Load-bearing premise

The system needs a noiseless enrollment utterance recorded from exactly the desired speaker's position with time-invariant acoustic paths, so the RTF ratio faithfully represents the target's spatial signature.

Editorial extensions

If this is right

  • On the random-location test set, the RTF-conditioned network reaches 9.2 dB SI-SDR and 0.81 STOI, above the DOA-conditioned variant (8.4 dB, 0.80) and the spectral-conditioned variant (8.18 dB, 0.80).
  • All three proposed spatial or spectral enrollment variants outperform the single-channel model and the estimated MVDR beamformer, suggesting that learned conditioning on enrollment cues can replace explicit beamforming for this task.
  • When two speakers share the same DOA, the RTF model still delivers 8.8 dB SI-SDR, on par with its different-DOA result and above the oracle MVDR's 8.5 dB, because the RTF distinguishes sources by distance and reverberant fingerprint rather than angle alone.
  • The method targets the reverberant replica of the desired speaker as received by the microphone array, matching the MVDR-style objective of estimating the desired signal at a reference microphone.
  • The RTF encoder's output is averaged over time to produce a single spatial representation, which lets the network condition the entire mixture embedding on the target's spatial signature.

Reading between the lines

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

  • If the RTF advantage survives noisy or position-mismatched enrollment, this design suggests that learned separators can consume spatial transfer functions directly, making explicit beamformer front-ends less necessary for target extraction.
  • A natural stress test is a moving speaker: the model assumes time-invariant acoustic transfer functions, so an online deployment would need to re-estimate or track the RTF from short clean segments.
  • The same conditioning mechanism could extend to more than two speakers or combine with visual cues, since the RTF is a per-position spatial fingerprint rather than a speaker-identity embedding.
  • The gap between TSE-RTF and TSE-DOA on the same-DOA trial hints that distance and reverberation information, not just angle, are what carry the improvement; quantifying how much each contributes would require an ablation that the paper does not include.
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 / 6 minor

Summary. The paper proposes a multi-microphone target speaker extraction (TSE) architecture that uses the instantaneous relative transfer function (RTF), estimated from a noiseless enrollment utterance recorded at the desired speaker's position, as a spatial enrollment cue. The authors compare this RTF-based cue against an oracle direction-of-arrival (DOA) cue and a single-channel spectral speaker embedding within the same end-to-end network, and also compare against oracle and estimated MVDR beamformers. Experiments on simulated reverberant mixtures with two speakers and directional noise report that the RTF-based variant achieves 9.2 dB SI-SDR on the random-location test set, outperforming the DOA-based variant (8.4 dB) and the spectral variant (8.18 dB). In an additional same-DOA scenario, the RTF-based model reaches 8.8 dB SI-SDR, and the paper concludes that RTF cues remain beneficial even when both speakers share the same direction.

Significance. If these results are reproducible and robust, the paper provides a useful practical comparison of enrollment cue types for end-to-end multichannel TSE, and the RTF-based spatial cue is a sensible alternative to DOA-based conditioning in reverberant conditions. The paper's main empirical contribution is the head-to-head comparison of RTF, DOA, and spectral cues under controlled simulation, and the finding that spatial cues outperform a purely spectral embedding is consistent with the broader beamforming and TSE literature. The same-DOA experiment is conceptually interesting because it goes beyond what a classical DOA-only beamformer can resolve. However, the strength of the conclusion currently rests on a small performance gap in a single test set without statistical or robustness analysis, and the same-DOA comparison omits the very DOA baseline needed to support its headline claim. The paper is clearly written and the experimental setup is mostly described in enough detail to be reproduced, but the missing baselines and lack of error bars prevent the current version from fully supporting the conclusions as stated.

major comments (3)
  1. [Section IV-E, Table I (bottom)] The same-DOA experiment reports only Oracle MVDR and TSE-RTF, yet the concluding claim is that RTF is beneficial 'even when both speakers originate from the same direction.' Without a same-DOA TSE-DOA result, the 8.8 dB SI-SDR of TSE-RTF cannot be attributed specifically to the RTF cue; the model might achieve comparable performance from spectral cues alone, or a DOA-conditioned model might also distinguish the speakers via distance-dependent magnitude differences. The paper's statement that a DOA model 'would fail to distinguish the speakers' is a hypothesis, not a measured baseline. Adding TSE-DOA and ideally TSE-Spectral results to the same-DOA block of Table I is necessary to support the paper's most distinctive claim.
  2. [Section IV-E and Table I] The central comparison between TSE-RTF (9.2 dB) and TSE-DOA (8.4 dB) is a 0.8 dB SI-SDR difference computed on only 300 test examples from a single training run, with no error bars, confidence intervals, paired significance tests, or multiple-seed results reported. Given the variability typical of deep speaker extraction models, this gap could plausibly be within run-to-run noise. The authors should report statistics across at least several independent training runs, or provide paired per-utterance significance testing, so that the superiority of RTF over DOA is quantitatively supported.
  3. [Section III-B.1 and Section IV-A] The proposed method relies on a noiseless enrollment signal recorded at exactly the desired speaker's position, and on the time-invariance of the acoustic transfer functions so that the instantaneous RTF ratio in Eq. (3) remains a valid spatial fingerprint. The paper acknowledges this setup but does not test its sensitivity to enrollment noise, small position mismatch, speaker movement, or RIR changes. Since the introduction states that 'in practice, any signal from the desired location can be utilized,' a robustness experiment with perturbed enrollment conditions is needed to establish whether the reported advantage over DOA holds under realistic enrollment imperfections.
minor comments (6)
  1. [Section III-A and Fig. 1] The architecture description omits several hyperparameters required for reproduction, such as the number of convolution channels, kernel sizes, the self-attention dimension, the number of heads, and the DOA lookup-table size; these should be listed in Section IV-B or in the figure caption.
  2. [Section III-B.1, Eq. (3)] The instantaneous RTF is defined as a per-frequency ratio of STFT coefficients, but no regularization or handling of near-zero denominator values is described; a brief note on the numerical implementation would clarify the feature extraction.
  3. [Table I] The table would be clearer if the random-location and same-DOA conditions were explicitly labeled as separate blocks with row headers, and if the same-DOA rows did not rely on the surrounding text to identify the experimental condition.
  4. [Throughout] The naming of the proposed variants is inconsistent: the text uses 'Proposed-RTF' in Section IV-E while the table and elsewhere use 'TSE-RTF'; please unify the terminology.
  5. [References] Reference [2] contains a typo: 'deparation' should be 'separation'.
  6. [Section IV-A] The sentence 'The utterances of the desired and interference sources are summed together with the reverberant noise and sensor noise' would be clearer if it explicitly stated that the same RIR used for the mixture is also used for the enrollment signal, although this is implied by the preceding paragraph.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the RTF-vs-DOA comparison is an empirical ablation with held-out data; the same-DOA claim lacks a DOA baseline but that is a completeness gap, not a circular step.

full rationale

The paper is an empirical comparison, not a formal derivation, so the circularity patterns (self-definitional, fitted-input-called-prediction, self-citation load-bearing, etc.) do not apply. The RTF feature is computed directly from the noiseless enrollment signal via Eq. (3) and fed into a neural network whose output is evaluated on a held-out test set (Table I), so no prediction reduces to its input by construction. The comparison against the authors' own single-channel system [18] is a baseline, not a load-bearing premise for the central RTF-vs-DOA claim. The self-citation of [30] (with overlapping authorship) is merely motivational and not used to justify the empirical result. One limitation is noted explicitly in Section IV-E: the same-DOA experiment evaluates only TSE-RTF and the Oracle MVDR, omitting the TSE-DOA baseline; consequently, the conclusion that RTF retains its benefit when speakers share a DOA is not directly supported by a measured comparison. However, this is a missing experimental condition, not a circular step. The core claim—RTF outperforms DOA and spectral cues on randomly located speakers—is an independent, data-driven result with external benchmarks (LibriSpeech, simulated RIRs), so the paper is self-contained against circularity.

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

The central claim is empirical and does not introduce ad hoc fitted constants in a derivation. The DNN weights are learned through supervised training, which is standard practice rather than a free parameter of the scientific claim. The relevant loaded assumptions are the availability of a clean, position-matched enrollment and the time-invariant two-source signal model.

assumptions (4)
  • domain assumption A noiseless enrollment signal is available from exactly the same position as the desired speaker.
    Section III-B.1 and Eq. (3) compute the instantaneous RTF as a ratio of microphone STFTs of this enrollment; Section IV-A explicitly uses a clean enrollment convolved with the same RIR. The paper does not test noisy or position-mismatched enrollment.
  • domain assumption Acoustic transfer functions are time-invariant within a recording.
    Eq. (1) defines h_q(k) as time-invariant ATFs relating each source to the microphone array. Moving speakers or time-varying channels would break the RTF estimate in Eq. (3).
  • domain assumption The evaluation scenario is limited to two concurrent speakers (Q=2) plus one directional noise source and sensor noise.
    Section II explicitly sets Q=2 and Section IV-A adds one directional noise source and pink sensor noise. Conclusions may not generalize to more speakers or multiple directional noise sources.
  • domain assumption Image-method simulated RIRs with uniformly random room dimensions and RT60 accurately represent reverberant environments.
    Section IV-A uses the Image Method to generate four-channel RIRs. All results are simulated, so real-world acoustic conditions are not tested.

how reviews work

0 comments
Cite this review

Pith. "Pith review of End-to-End Multi-Microphone Speaker Extraction Using Relative Transfer Functions." pith.science (2026). https://pith.science/paper/5S6IACKB

@misc{pith2026250206285,
  author       = {Pith},
  title        = {Pith review of: End-to-End Multi-Microphone Speaker Extraction Using Relative Transfer Functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5S6IACKB}},
  note         = {Machine review of arXiv:2502.06285}
}
read the original abstract

This paper introduces a multi-microphone method for extracting a desired speaker from a mixture involving multiple speakers and directional noise in a reverberant environment. In this work, we propose leveraging the instantaneous relative transfer function (RTF), estimated from a reference utterance recorded in the same position as the desired source. The effectiveness of the RTF-based spatial cue is compared with direction of arrival (DOA)-based spatial cue and the conventional spectral embedding. Experimental results in challenging acoustic scenarios demonstrate that using spatial cues yields better performance than the spectral-based cue and that the instantaneous RTF outperforms the DOA-based spatial cue.

Figures

Figures reproduced from arXiv: 2502.06285 by the authors.

Figure 1
Figure 1. The multichannel TSE algorithm is described, comparing three alternative enrollment features: (1) the instantaneous RTF extracted from clean speech [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Spectrogram example of the Proposed-RTF model compared with the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 34 canonical work pages

  1. [1]

    FaSNet: Low- latency adaptive beamforming for multi-microphone audio processing,

    Y . Luo, C. Han, N. Mesgarani, E. Ceolini, and S.-C. Liu, “FaSNet: Low- latency adaptive beamforming for multi-microphone audio processing,” in IEEE Workshop Autom. Speech Recog. and Underst. (ASRU) , 2019, pp. 260–267

    work page 2019

  2. [2]

    Implicit filter-and-sum network for end-to- end multi-channel speech deparation,

    Y . Luo and N. Mesgarani, “Implicit filter-and-sum network for end-to- end multi-channel speech deparation,” in ISCA Interspeech , 2021, pp. 3071–3075

    work page 2021

  3. [3]

    Implicit neural spatial filtering for multichannel source separation in the waveform domain,

    D. Markovic, A. Defossez, and A. Richard, “Implicit neural spatial filtering for multichannel source separation in the waveform domain,” in ISCA Interspeech, 2022, pp. 1806–1810

    work page 2022

  4. [4]

    ADL- MVDR: All deep learning MVDR beamformer for target speech sepa- ration,

    Z. Zhang, Y . Xu, M. Yu, S.-X. Zhang, L. Chen, and D. Yu, “ADL- MVDR: All deep learning MVDR beamformer for target speech sepa- ration,” in IEEE Int. Conf. Acoustics, Speech, Signal Proc. (ICASSP) , 2021, pp. 6089–6093

    work page 2021

  5. [5]

    Combining deep neural networks and beamforming for real-time multi-channel speech enhancement using a wireless acoustic sensor network,

    E. Ceolini and S.-C. Liu, “Combining deep neural networks and beamforming for real-time multi-channel speech enhancement using a wireless acoustic sensor network,” in IEEE Int. Workshop on Machine Learning for Signal Proc. (MLSP) , 2019

    work page 2019

  6. [6]

    NICE-beam: Neural integrated covariance estimators for time-varying beamformers,

    J. Casebeer, J. Donley, D. Wong, B. Xu, and A. Kumar, “NICE-beam: Neural integrated covariance estimators for time-varying beamformers,” arXiv:2112.04613, 2021

    arXiv 2021

  7. [7]

    A causal U-Net based neural beamforming network for real-time multi- channel speech enhancement,

    X. Ren, X. Zhang, L. Chen, X. Zheng, C. Zhang, L. Guo, and B. Yu, “A causal U-Net based neural beamforming network for real-time multi- channel speech enhancement,” in Interspeech, 2021, pp. 1832–1836

    work page 2021

  8. [8]

    Embedding and beamforming: All-neural causal beamformer for multichannel speech enhancement,

    A. Li, W. Liu, C. Zheng, and X. Li, “Embedding and beamforming: All-neural causal beamformer for multichannel speech enhancement,” in IEEE Int. Conf. Acoustics, Speech, Signal Proc. (ICASSP) , 2022, pp. 6487–6491

    work page 2022

Show all 36 references
  1. [9]

    Complex-valued spatial au- toencoders for multichannel speech enhancement,

    M. M. Halimeh and W. Kellermann, “Complex-valued spatial au- toencoders for multichannel speech enhancement,” in IEEE Int. Conf. Acoustics, Speech, Signal Proc. (ICASSP) , 2022, pp. 261–265

    2022

  2. [10]

    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 Int. Conf. Acoustics, Speech, Signal Proc. (ICASSP), 2023

    2023

  3. [11]

    Explainable DNN-based beamformer with postfilter,

    A. Cohen, D. Wong, J.-S. Lee, and S. Gannot, “Explainable DNN-based beamformer with postfilter,” IEEE/ACM Trans. Audio, Speech, Language Proc., Jul. 2024, submitted

    2024

  4. [12]

    Multi- microphone simultaneous speakers detection and localization of multi- sources for separation and noise reduction,

    A. Schwartz, O. Schwartz, S. E. Chazan, and S. Gannot, “Multi- microphone simultaneous speakers detection and localization of multi- sources for separation and noise reduction,” EURASIP J. on Audio, Speech, and Music Proc. , vol. 2024, no. 1, p. 50, 2024

    2024

  5. [13]

    Neural target speech extraction: An overview,

    K. ˇZmol´ıkov´a, M. Delcroix, T. Ochiai, K. Kinoshita, J. ˇCernock´y, and D. Yu, “Neural target speech extraction: An overview,” IEEE Signal Proc. Magazine, vol. 40, no. 3, pp. 8–29, 2023

    2023

  6. [14]

    Single channel target speaker extraction and recognition with speaker beam,

    M. Delcroix, K. ˇZmol´ıkov´a, K. Kinoshita, A. Ogawa, and T. Nakatani, “Single channel target speaker extraction and recognition with speaker beam,” in IEEE Int. Conf. Acoustics, Speech, Signal Proc. (ICASSP) , 2018, pp. 5554–5558

    2018

  7. [15]

    Time-domain speaker extraction network,

    C. Xu, W. Rao, E. S. Chng, and H. Li, “Time-domain speaker extraction network,” in IEEE Workshop Autom. Speech Recog. and Underst. (ASRU), 2019, pp. 327–334

    2019

  8. [16]

    Improving speaker discrimination of tar- get speech extraction with time-domain speakerbeam,

    M. Delcroix, T. Ochiai, K. Zmolikova, K. Kinoshita, N. Tawara, T. Nakatani, and S. Araki, “Improving speaker discrimination of tar- get speech extraction with time-domain speakerbeam,” in IEEE In- ternational Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020...

    2020

  9. [17]

    Single microphone speaker extraction using unified time-frequency Siamese-Unet,

    A. Eisenberg, S. Gannot, and S. E. Chazan, “Single microphone speaker extraction using unified time-frequency Siamese-Unet,” in Proc. Euro. Signal Proc. Conf. (EUSIPCO) , Aug. 2022, pp. 762–766

    2022

  10. [18]

    A two-stage speaker extraction algorithm under adverse acoustic conditions using a single-microphone,

    ——, “A two-stage speaker extraction algorithm under adverse acoustic conditions using a single-microphone,” in Proc. Euro. Signal Proc. Conf. (EUSIPCO), 2023, pp. 266–270

    2023

  11. [19]

    Learning speaker representation for neural network based multichannel speaker extraction,

    K. ˇZmol´ıkov´a, M. Delcroix, K. Kinoshita, T. Higuchi, A. Ogawa, and T. Nakatani, “Learning speaker representation for neural network based multichannel speaker extraction,” in IEEE Automatic Speech Recognition and Understanding Workshop (ASRU) , 2017, pp. 8–15

    2017

  12. [20]

    An investigation into the multi- channel time domain speaker extraction network,

    C. Zoril ˘a, M. Li, and R. Doddipatla, “An investigation into the multi- channel time domain speaker extraction network,” in IEEE Spoken Language Technology Workshop (SLT) , 2021, pp. 793–800

    2021

  13. [21]

    Multi-channel target speech extraction with channel decorrelation and target speaker adaptation,

    J. Han, X. Zhou, Y . Long, and Y . Li, “Multi-channel target speech extraction with channel decorrelation and target speaker adaptation,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021, pp. 6094–6098

    2021

  14. [22]

    L-spex: Localized target speaker extraction,

    M. Ge, C. Xu, L. Wang, E. S. Chng, J. Dang, and H. Li, “L-spex: Localized target speaker extraction,” in IEEE Int. Conf. Acoustics, Speech, Signal Proc. (ICASSP) , 2022, pp. 7287–7291

    2022

  15. [23]

    Beamformer- guided target speaker extraction,

    M. Elminshawi, S. R. Chetupalli, and E. A. Habets, “Beamformer- guided target speaker extraction,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2023, pp. 1–5

    2023

  16. [24]

    Multi-modal multi-channel target speech separation,

    R. Gu, S.-X. Zhang, Y . Xu, L. Chen, Y . Zou, and D. Yu, “Multi-modal multi-channel target speech separation,” IEEE Journal of Selected Topics in Signal Processing , vol. 14, no. 3, pp. 530–541, 2020

    2020

  17. [25]

    Neural spatio-temporal beamformer for target speech separation,

    Y . Xu, M. Yu, S.-X. Zhang, L. Chen, C. Weng, J. Liu, and D. Yu, “Neural spatio-temporal beamformer for target speech separation,” arXiv preprint arXiv:2005.03889, 2020

    2005 arXiv

  18. [26]

    Generalized spatio- temporal rnn beamformer for target speech separation,

    Y . Xu, Z. Zhang, M. Yu, S.-X. Zhang, and D. Yu, “Generalized spatio- temporal rnn beamformer for target speech separation,” arXiv preprint arXiv:2101.01280, 2021

    2021 arXiv

  19. [27]

    Spatially selective deep non-linear filters for speaker extraction,

    K. Tesch and T. Gerkmann, “Spatially selective deep non-linear filters for speaker extraction,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2023, pp. 1–5

    2023

  20. [28]

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

    ——, “Insights into deep non-linear filters for improved multi-channel speech enhancement,” IEEE/ACM Trans. Audio, Speech, Language Proc., vol. 31, pp. 563–575, 2023

    2023

  21. [29]

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

    ——, “Multi-channel speech separation using spatially selective deep non-linear filters,” IEEE/ACM Trans. Audio, Speech, Language Proc. , vol. 32, pp. 542–553, 2024

    2024

  22. [30]

    A consol- idated perspective on multimicrophone speech enhancement and source separation,

    S. Gannot, E. Vincent, S. Markovich-Golan, and A. Ozerov, “A consol- idated perspective on multimicrophone speech enhancement and source separation,” IEEE/ACM Trans. Audio, Speech, Language Proc. , vol. 25, no. 4, pp. 692–730, Apr. 2017

    2017

  23. [31]

    Signal enhancement using beamforming and nonstationarity with applications to speech,

    S. Gannot, D. Burshtein, and E. Weinstein, “Signal enhancement using beamforming and nonstationarity with applications to speech,” IEEE Transactions on Signal Processing , vol. 49, no. 8, pp. 1614–1626, 2001

    2001

  24. [32]

    Multi-microphone speaker separation based on deep doa estimation,

    S. E. Chazan, H. Hammer, G. Hazan, J. Goldberger, and S. Gannot, “Multi-microphone speaker separation based on deep doa estimation,” in Proc. Euro. Signal Proc. Conf. (EUSIPCO) , 2019

    2019

  25. [33]

    Room impulse response generator,

    E. A. Habets, “Room impulse response generator,” Technische Univer- siteit Eindhoven, Tech. Rep , vol. 2, no. 2.4, 2006

    2006

  26. [34]

    Adam: A method for stochastic optimization,

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” in International Conference on Learning Representations (ICLR) , 2015

    2015

  27. [35]

    An algorithm for intelligibility prediction of time-frequency weighted noisy speech,

    C. H. Taal, R. C. Hendriks, R. Heusdens, and J. Jensen, “An algorithm for intelligibility prediction of time-frequency weighted noisy speech,” IEEE Trans. Audio, Speech, Language Proc. , vol. 19, no. 7, pp. 2125– 2136, 2011

    2011

  28. [36]

    Performance analysis of the covariance-whitening and the covariance-subtraction methods for estimating the relative transfer function,

    S. Markovich-Golan, S. Gannot, and W. Kellermann, “Performance analysis of the covariance-whitening and the covariance-subtraction methods for estimating the relative transfer function,” in Proc. Euro. Signal Proc. Conf. (EUSIPCO) , Rome, Italy, Sep. 2018

    2018

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.