Pith. sign in

REVIEW 3 major objections 4 minor 35 references

Multi-Utterance Speech Separation and Association Trained on Short Segments

T0 review · 3 major / 4 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 A frequency-temporal RNN trained on 10-second segments reports direct separation of 21-121 second two-speaker mixtures while preserving each speaker's utterances in one output stream across 40-second silences.

desk verdict Useful empirical study with a real generalization result, but the direct-vs-stitching comparison is weakened by a segment-length mismatch and an underspecified DER metric. read the letter →

arxiv 2507.02562 v1 pith:QC77QTUK submitted 2025-07-03 eess.AS cs.SD

Yuzhu Wang , Archontis Politis , Konstantinos Drossos , Tuomas Virtanen This is my paper · ORCID
classification eess.AScs.SD
keywords speechseparationmulti-utterancespeakerassociationlong-formaudiofrequency-temporalrecurrentnetworkpermutationinvarianttrainingutterancegapgeneralizationsingle-channel
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 argues that a speech separation model trained on short, fixed-length segments can be applied directly to much longer multi-utterance recordings without segmentation or stitching. The proposed frequency-temporal recurrent neural network (FTRNN) is trained on 10-second mixtures using permutation-invariant SI-SDR loss, yet on generated 21-121 second two-speaker mixtures it reports 15.2 dB SI-SDR and 6.9% diarization error rate, beating a strong stitching baseline that is given perfect permutation information. The result matters because current practice splits long audio into short segments, separates each, and stitches, which distorts boundaries and needs external speaker permutation logic. If the claim holds, long-form separation can be done by a single lightweight recurrent model that keeps each speaker's utterances in one output stream across silences longer than any seen in training.

What carries the argument

The load-bearing mechanism is the FTRNN's paired recurrent modules: an along-frequency full-band bidirectional LSTM models dependencies across frequency bins within each time frame, and an along-temporal sub-band bidirectional LSTM models time dynamics independently in each frequency bin. These modules are stacked in four residual blocks. Because the temporal BLSTM sees the whole unsegmented input at inference, its hidden state can carry speaker identity across long silences without needing explicit speaker-ID or clustering modules. Training uses permutation invariant training with SI-SDR, and no cross-segment permutation consistency is enforced, so the association across gaps emerges from the recurrent processing rather than from stitching logic.

What would settle it

Run the described FTRNN on two-speaker mixtures with a 40-second silent gap between a speaker's utterances, convert each separated stream to speaker-activity hypotheses with an energy-based VAD that has no access to ground-truth boundaries, and compute DER; if DER rises far above the reported single-digit values, the long-gap association claim is disproved.

Watch

Extended reading notes

Core claim

The paper's central claim is that a recurrent model with full-band and sub-band processing can bridge the train-short/infer-long gap in speech separation. Trained only on 10-second segments, FTRNN performs inference on unsegmented mixtures up to 121 seconds and maintains output-stream speaker association across utterance gaps of 40 seconds, although training gaps were only 1-3 seconds. In the paper's experiments it reports 15.2 dB SI-SDR and 6.9% DER by direct inference on the main test set, exceeding every baseline, including TFGrid with oracle stitching; and with oracle stitching it still surpasses that baseline. The authors interpret this as evidence that direct long-sequence inference captures context that segment-stitch pipelines lose.

Load-bearing premise

The speaker-association claim rests on DER scores for the separated streams, but the paper does not state how waveforms are converted into hypothesis speaker-activity segments; if that conversion used the known utterance boundaries, the reported stability across 40-second gaps would be an artifact of the evaluation rather than a true property of the model.

Editorial extensions

If this is right

  • Long-audio speech separation no longer requires a segmentation and stitching pipeline; a model trained on 10-second clips can process 21-121 second mixtures directly.
  • Direct inference can outperform oracle-stitching, so the segment-stitch paradigm's best-case performance is not an upper bound on what a single-stream model can achieve.
  • Utterance gap duration, not utterance count, is the dominant difficulty for multi-utterance separation and association; models degrade more when silence between utterances grows from 3 s to 40 s than when utterances per speaker grow from one to five.
  • Speaker consistency across output streams can be achieved by recurrence alone, with 0.9 M parameters, without speaker embeddings, clustering, or attractor modules.
  • Replacing SI-SDR with SA-SDR lowered the result to 12.2 dB SI-SDR, indicating the training objective affects long-form association performance.

Reading between the lines

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

  • Because the temporal BLSTM operates on the full unsegmented input, the method should extend to even longer recordings up to memory and compute limits; a natural test is whether the small gap-robustness holds for 5-10 minute inputs.
  • The absence of a VAD or post-processing description in the DER evaluation means the speaker-association result should be re-tested with a standard energy-based VAD on separated streams; if oracle boundaries were used, the 40-second-gap robustness may shrink.
  • The architecture suggests a streaming variant: replacing the bidirectional temporal LSTM with a causal one would trade some gap-robustness for online operation, which matters for live meeting and telephony applications.
  • The observed gap-duration effect implies that datasets should be built with long silences between utterances, not just more utterances, if models are to generalize to natural conversations.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper addresses the problem of multi-utterance speech separation when models trained on short (10 s) segments are applied to longer recordings. The authors propose FTRNN, a lightweight recurrent architecture with full-band and sub-band BLSTM modules, trained with PIT and SI-SDR. They evaluate FTRNN and several baselines (DPRNN, DPTNet, SepFormer, TFGrid) on a synthetic multi-utterance dataset derived from LibriSpeech and DEMAND, with test sets varying utterance gaps (3-40 s) and utterance counts (1-5). They compare direct inference on unsegmented long audio with an oracle-stitching paradigm (using ground-truth to choose the best permutation per segment). The main claims are that FTRNN generalizes to longer inputs, preserves speaker association across larger gaps than seen in training, and that direct inference outperforms the conventional segment-stitch approach.

Significance. If validated, the paper would make a useful contribution: it demonstrates a 0.9 M-parameter model that can process 21-121 s mixtures without segmentation, outperforming the tested baselines, including the strong TFGrid baseline under oracle stitching. The controlled experimental design (retrained baselines, multiple gap/utterance settings, oracle-stitching upper bound) is a strength, and the SI-SDR stability across utterance gaps is a notable finding. However, the load-bearing comparisons are currently undermined by the stitching segment-length mismatch and an unspecified DER protocol, so the headline claims are not yet fully supported.

major comments (3)
  1. [Section 5 (stitching setup)] The oracle stitching evaluation divides long recordings into 5 s segments with 20 % overlap for all models, whereas FTRNN, DPRNN, and DPTNet are trained on 10 s segments (only TFGrid is trained on 5 s). This mismatch reduces the context available to the stitching-based baselines and can inflate the apparent advantage of direct inference. The claim in Section 5.1 that 'processing longer recordings directly is more effective than the segment-stitch approach' is therefore not conclusively supported; indeed, Fig. 3 shows that FTRNN with oracle stitching at a 3 s gap reaches 16.4 dB, above its direct-inference 15.8 dB. Please repeat the oracle-stitching evaluation with a segment length matched to the training length (10 s for FTRNN, DPRNN, and DPTNet; 5 s for TFGrid) and report whether the direct-inference advantage persists.
  2. [Section 4.2 (evaluation metrics)] The computation of the diarization error rate (DER) is not specified. The paper states only that DER is used 'to assess speaker activity and association', but does not say how the continuous separated waveforms are converted into hypothesis speaker-activity segments (e.g., VAD threshold, minimum duration, collar size, or whether oracle utterance/silence boundaries are used). Without this protocol, the DER values in Table 2 and any claim about speaker association across utterance gaps cannot be verified or compared with other systems. Please provide the complete DER evaluation pipeline, including all thresholds and any use of reference boundaries.
  3. [Section 5 (all results)] All reported results are single-point estimates without error bars, standard deviations, confidence intervals, or significance tests. The paper repeatedly uses 'significant' (e.g., Sections 1 and 5.3) and draws conclusions from small differences such as the 0.4 dB improvement over TFGrid oracle stitching in Table 2, which could be within run-to-run or test-sample variability. Please report variance across test samples or training runs and, where appropriate, paired significance tests for the key comparisons.
minor comments (4)
  1. [Section 5.1] The sentence 'DPRNN offers the lowest parameter count at 2.6 M' is incorrect, because FTRNN has 0.9 M parameters; it should read 'among the baselines' or be revised to name FTRNN.
  2. [Section 6] The conclusion that 'direct inference on long signals outperformed segment-separation-stitch results even with ideal permutation information' is not consistent with Fig. 3, where FTRNN oracle stitching at a 3 s gap (16.4 dB) exceeds direct inference (15.8 dB). The conclusion should be qualified as applying to the overall test configuration, not universally.
  3. [Table 2] The superscript footnote markers after model names (e.g., 'DPRNN2') are easy to misread as model variants; consider placing footnote numbers after the reference citations instead.
  4. [Section 4.1] The synthetic evaluation is well controlled, but all test sets are generated with the same simulation pipeline as training; a brief discussion of the potential gap to real recordings would strengthen the paper's claims of practical applicability.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation: FTRNN is an empirical architecture trained with an external loss on held-out data, and the only same-author citation is not load-bearing.

full rationale

The paper's central claims are empirical rather than derivational: an FTRNN trained with permutation-invariant training and the SI-SDR loss on 10 s segments is applied directly to 21-121 s mixtures and compared with segment-stitch baselines. The oracle-stitching baseline uses ground-truth reference signals only to choose the optimal SI-SDR permutation, which is an upper bound for the stitching paradigm and is not an input to FTRNN; the direct-versus-stitch comparison is therefore conservative rather than circular. Generalization across utterance gaps is measured on held-out generated test sets whose speakers and noise signals do not overlap with training, and the training objective is a standard external metric. The same-author citation [26] appears only in a general list of attractor-based related work and is not used to justify the architecture, the loss, or the evaluation, so it is not load-bearing. Two reporting gaps exist but are not circularity: Section 5 uses 5 s stitching segments for models trained on 10 s segments, which is an evaluation-fairness concern about the headline comparison, and Section 4.2 does not specify how continuous separated waveforms are converted into DER hypothesis segments, which is an evaluation-specification gap. Neither gap makes a prediction equivalent to an input by construction. Accordingly, no circular step is identified and the circularity score is low.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central empirical claim rests on the synthetic data generation pipeline being representative, on the two-speaker concatenated-utterance signal model, and on an unspecified DER conversion step. No invented entities are introduced. The model hyperparameters are hand-tuned on validation data and are listed as free parameters.

free parameters (1)
  • Architecture hyperparameters (D, N, BLSTM hidden units) = D=32, N=4, 96 hidden units per direction
    Determined through pre-experiments on the validation set, as stated in Section 4.2, rather than derived from first principles. They affect the generalization result and are hand-chosen.
assumptions (3)
  • domain assumption The LibriSpeech plus DEMAND plus gpuRIR generation pipeline represents the target long multi-utterance separation conditions.
    All training, validation, and ten test subsets come from this synthetic pipeline in Section 4.1. No real recordings are used, so real-world distribution shift is not measured.
  • domain assumption The two-speaker signal model with concatenated utterances and silence gaps is the intended target scenario.
    The method and evaluation assume C=2 and non-overlapping own-utterance silences. Extension to more speakers or overlapping same-speaker speech is left to future work.
  • domain assumption Diarization error rate computed by pyannote.metrics on the separated outputs is a valid measure of speaker association.
    The paper uses DER to assess association but does not specify the VAD or thresholding used to convert continuous separation estimates into diarization hypotheses, as noted in Section 4.2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multi-Utterance Speech Separation and Association Trained on Short Segments." pith.science (2026). https://pith.science/paper/QC77QTUK

@misc{pith2026250702562,
  author       = {Pith},
  title        = {Pith review of: Multi-Utterance Speech Separation and Association Trained on Short Segments},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QC77QTUK}},
  note         = {Machine review of arXiv:2507.02562}
}
read the original abstract

Current deep neural network (DNN) based speech separation faces a fundamental challenge -- while the models need to be trained on short segments due to computational constraints, real-world applications typically require processing significantly longer recordings with multiple utterances per speaker than seen during training. In this paper, we investigate how existing approaches perform in this challenging scenario and propose a frequency-temporal recurrent neural network (FTRNN) that effectively bridges this gap. Our FTRNN employs a full-band module to model frequency dependencies within each time frame and a sub-band module that models temporal patterns in each frequency band. Despite being trained on short fixed-length segments of 10 s, our model demonstrates robust separation when processing signals significantly longer than training segments (21-121 s) and preserves speaker association across utterance gaps exceeding those seen during training. Unlike the conventional segment-separation-stitch paradigm, our lightweight approach (0.9 M parameters) performs inference on long audio without segmentation, eliminating segment boundary distortions while simplifying deployment. Experimental results demonstrate the generalization ability of FTRNN for multi-utterance speech separation and speaker association.

Figures

Figures reproduced from arXiv: 2507.02562 by the authors.

Figure 1
Figure 1. Visualization of long speech separation: (a) training phase [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Proposed system architecture. Both modules employ BLSTM networks, which are key to the model’s ability to process signals longer than the segments used during training. The two modules are arranged sequentially with residual connections and layer normalization, repeated N times. Finally, a 2D deconvolutional layer maps the processed features back to the time-frequency domain with 2C output channels, where C is the n… view at source ↗
Figure 3
Figure 3. SI-SDR performance comparison across different utterance [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 35 canonical work pages

  1. [1]

    Vincent, T

    E. Vincent, T. Virtanen, and S. Gannot, Audio Source Separation and Speech Enhancement. John Wiley & Sons, 2018

    work page 2018

  2. [2]

    Deep learning for audio signal processing,

    H. Purwins, B. Li, T. Virtanen, J. Schl ¨uter, S.-Y . Chang, and T. Sainath, “Deep learning for audio signal processing,” IEEE Journal of Selected Topics in Signal Processing , vol. 13, no. 2, pp. 206–219, 2019

    work page 2019

  3. [3]

    An overview of machine learning and other data-based methods for spatial audio capture, processing, and reproduction,

    M. Cobos, J. Ahrens, K. Kowalczyk, and A. Politis, “An overview of machine learning and other data-based methods for spatial audio capture, processing, and reproduction,” EURASIP Journal on Audio, Speech, and Music Processing, vol. 2022, no. 1, p. 10, 2022

    work page 2022

  4. [4]

    Supervised speech separation based on deep learning: An overview,

    D. Wang and J. Chen, “Supervised speech separation based on deep learning: An overview,” IEEE/ACM Trans. Audio, Speech, Lang. Process., vol. 26, no. 10, pp. 1702–1726, 2018

    work page 2018

  5. [5]

    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 Trans. Audio, Speech, Lang. Process., vol. 27, no. 8, pp. 1256–1266, 2019

    work page 2019

  6. [6]

    Permutation invariant training of deep models for speaker-independent multi-talker speech separation,

    D. Yu, M. Kolbæk, Z.-H. Tan, and J. Jensen, “Permutation invariant training of deep models for speaker-independent multi-talker speech separation,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. , 2017, pp. 241–245

    work page 2017

  7. [7]

    Multitalker speech separation with utterance-level permutation invariant training of deep recurrent neural networks,

    M. Kolbæk, D. Yu, Z.-H. Tan, and J. Jensen, “Multitalker speech separation with utterance-level permutation invariant training of deep recurrent neural networks,” IEEE/ACM Trans. Audio, Speech, Lang. Process., vol. 25, no. 10, pp. 1901–1913, 2017

    work page 1901

  8. [8]

    TF-GridNet: Making time-frequency domain models great again for monaural speaker separation,

    Z.-Q. Wang, S. Cornell, S. Choi, Y . Lee, B.-Y . Kim, and S. Watanabe, “TF-GridNet: Making time-frequency domain models great again for monaural speaker separation,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process., 2023, pp. 1–5

    work page 2023

Show all 35 references
  1. [9]

    TF-GridNet: Integrating full- and sub-band modeling for speech separation,

    ——, “TF-GridNet: Integrating full- and sub-band modeling for speech separation,” IEEE/ACM Trans. Audio, Speech, Lang. Process. , vol. 31, pp. 3221–3236, 2023

    2023

  2. [10]

    Deep Clustering: Discriminative embeddings for segmentation and separation,

    J. R. Hershey, Z. Chen, J. Le Roux, and S. Watanabe, “Deep Clustering: Discriminative embeddings for segmentation and separation,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. , 2016, pp. 31–35

    2016

  3. [11]

    Low-latency deep clustering for speech separation,

    S. Wang, G. Naithani, and T. Virtanen, “Low-latency deep clustering for speech separation,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process., 2019, pp. 76–80

    2019

  4. [12]

    Wavesplit: End-to-end speech separation by speaker clustering,

    N. Zeghidour and D. Grangier, “Wavesplit: End-to-end speech separation by speaker clustering,” IEEE/ACM Trans. Audio, Speech, Lang. Process. , vol. 29, pp. 2840–2849, 2021

    2021

  5. [13]

    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 Proc. IEEE Int. Conf. Acoust., Speech Signal Process. IEEE, 2020, pp. 7284–7288

    2020

  6. [14]

    Dual-path modeling for long recording speech separation in meetings,

    C. Li, Z. Chen, Y . Luo, C. Han, T. Zhou, K. Kinoshita, M. Delcroix, S. Watanabe, and Y . Qian, “Dual-path modeling for long recording speech separation in meetings,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. IEEE, 2021, pp. 5739–5743

    2021

  7. [15]

    Continuous speech sep- aration using speaker inventory for long recording

    C. Han, Y . Luo, C. Li, T. Zhou, K. Kinoshita, S. Watanabe, M. Delcroix, H. Erdogan, J. R. Hershey, N. Mesgarani et al., “Continuous speech sep- aration using speaker inventory for long recording.” in Proc. Interspeech, 2021, pp. 3036–3040

    2021

  8. [16]

    Dual-path rnn for long recording speech separation,

    C. Li, Y . Luo, C. Han, J. Li, T. Yoshioka, T. Zhou, M. Delcroix, K. Kinoshita, C. Boeddeker, Y . Qian et al. , “Dual-path rnn for long recording speech separation,” in Proc. SLT. IEEE, 2021, pp. 865–872

    2021

  9. [17]

    Dual-Path RNN: Efficient long sequence modeling for time-domain single-channel speech separation,

    Y . Luo, Z. Chen, and T. Yoshioka, “Dual-Path RNN: Efficient long sequence modeling for time-domain single-channel speech separation,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. , 2020, pp. 46–50

    2020

  10. [18]

    Segment-less continuous speech separation of meetings: Training and evaluation criteria,

    T. von Neumann, K. Kinoshita, C. Boeddeker, M. Delcroix, and R. Haeb- Umbach, “Segment-less continuous speech separation of meetings: Training and evaluation criteria,” IEEE/ACM Trans. Audio, Speech, Lang. Process., vol. 31, pp. 576–589, 2022

    2022

  11. [19]

    PLDA for speaker verification with utterances of arbitrary duration,

    P. Kenny, T. Stafylakis, P. Ouellet, M. J. Alam, and P. Dumouchel, “PLDA for speaker verification with utterances of arbitrary duration,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. IEEE, 2013, pp. 7649–7653

    2013

  12. [20]

    X- vectors: Robust dnn embeddings for speaker recognition,

    D. Snyder, D. Garcia-Romero, G. Sell, D. Povey, and S. Khudanpur, “X- vectors: Robust dnn embeddings for speaker recognition,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. IEEE, 2018, pp. 5329–5333

    2018

  13. [21]

    Speaker recognition for multi-speaker conversations using x-vectors,

    D. Snyder, D. Garcia-Romero, G. Sell, A. McCree, D. Povey, and S. Khudanpur, “Speaker recognition for multi-speaker conversations using x-vectors,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. IEEE, 2019, pp. 5796–5800

    2019

  14. [22]

    Deep attractor network for single- microphone speaker separation,

    Z. Chen, Y . Luo, and N. Mesgarani, “Deep attractor network for single- microphone speaker separation,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process., 2017, pp. 246–250

    2017

  15. [23]

    Speaker-independent speech separation with deep attractor network,

    Y . Luo, Z. Chen, and N. Mesgarani, “Speaker-independent speech separation with deep attractor network,” IEEE/ACM Trans. Audio, Speech, Lang. Process., vol. 26, no. 4, pp. 787–796, 2018

    2018

  16. [24]

    Speech separation for an unknown number of speakers using transformers with encoder-decoder attractors,

    S. R. Chetupalli and E. A. P. Habets, “Speech separation for an unknown number of speakers using transformers with encoder-decoder attractors,” in Proc. Interspeech, 2022, pp. 5393–5397

    2022

  17. [25]

    Boosting unknown-number speaker separation with transformer decoder-based attractor,

    Y . Lee, S. Choi, B.-Y . Kim, Z.-Q. Wang, and S. Watanabe, “Boosting unknown-number speaker separation with transformer decoder-based attractor,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. , 2024, pp. 446–450

    2024

  18. [26]

    Attractor-based speech separation of multiple utterances by unknown number of speakers,

    Y . Wang, A. Politis, K. Drossos, and T. Virtanen, “Attractor-based speech separation of multiple utterances by unknown number of speakers,” arXiv:2505.16607, 2025

    2025 arXiv

  19. [27]

    SDR – half- baked or well done?

    J. Le Roux, S. Wisdom, H. Erdogan, and J. R. Hershey, “SDR – half- baked or well done?” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process., 2019, pp. 626–630

    2019

  20. [28]

    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 Proc. IEEE Int. Conf. Acoust., Speech Signal Process. , 2015

    2015

  21. [29]

    The diverse environments multi- channel acoustic noise database (DEMAND): A database of multichannel environmental noise recordings,

    J. Thiemann, N. Ito, and E. Vincent, “The diverse environments multi- channel acoustic noise database (DEMAND): A database of multichannel environmental noise recordings,” in Proceedings of Meetings on Acoustics, vol. 19, no. 1. AIP Publishing, 2013

    2013

  22. [30]

    gpuRIR: A python library for room impulse response simulation with gpu acceleration,

    D. Diaz-Guerra, A. Miguel, and J. R. Beltran, “gpuRIR: A python library for room impulse response simulation with gpu acceleration,” Multimedia Tools and Applications, vol. 80, pp. 5653–5671, 2021

    2021

  23. [31]

    ESPnet: End-to-end speech processing toolkit,

    S. Watanabe, T. Hori, S. Karita, T. Hayashi, J. Nishitoba, Y . Unno, N. En- rique Yalta Soplin, J. Heymann, M. Wiesner, N. Chen, A. Renduchintala, and T. Ochiai, “ESPnet: End-to-end speech processing toolkit,” in Proc. Interspeech, 2018, pp. 2207–2211

    2018

  24. [32]

    Pyannote. metrics: A toolkit for reproducible evaluation, diagnostic, and error analysis of speaker diarization systems,

    H. Bredin, “Pyannote. metrics: A toolkit for reproducible evaluation, diagnostic, and error analysis of speaker diarization systems,” in Proc. Interspeech, 2017, pp. 3587–3591

    2017

  25. [33]

    Dual-path transformer network: Direct context-aware modeling for end-to-end monaural speech separation,

    J. Chen, Q. Mao, and D. Liu, “Dual-path transformer network: Direct context-aware modeling for end-to-end monaural speech separation,” in Proc. Interspeech, 2020, pp. 2642–2646

    2020

  26. [34]

    Attention is all you need in speech separation,

    C. Subakan, M. Ravanelli, S. Cornell, M. Bronzi, and J. Zhong, “Attention is all you need in speech separation,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. , 2021, pp. 21–25

    2021

  27. [35]

    SA-SDR: A novel loss function for separation of meeting style data,

    T. von Neumann, K. Kinoshita, C. Boeddeker, M. Delcroix, and R. Haeb- Umbach, “SA-SDR: A novel loss function for separation of meeting style data,” in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. IEEE, 2022, pp. 6022–6026

    2022

Pith tools

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