Pith. sign in

REVIEW 4 major objections 7 minor 55 references

Classical Guitar Duet Separation using GuitarDuets -- a Dataset of Real and Synthesized Guitar Recordings

T0 review · 4 major / 7 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 Training on both real and synthesized classical guitar duets improves same-timbre separation more than training on either data type alone.

desk verdict The GuitarDuets dataset is a real and useful contribution, but the abstract's claim that combining real and synthetic data improves separation is only supported for guitar 1's SDR, and the lack of error bars means even that gain is shaky. read the letter →

arxiv 2507.01172 v1 pith:5HZTNI4N submitted 2025-07-01 eess.AS

Marios Glytsos , Christos Garoufis , Athanasia Zlatintsi , Petros Maragos This is my paper · ORCID
classification eess.AS
keywords musicsourceseparationmonotimbralclassicalguitarduetDuetsdatasetDemucsscore-informedsyntheticaudiopermutation-invarianttraining
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

Classical guitar duets are hard to separate because both instruments share a similar timbre, so standard source separation models struggle to tell them apart. The paper introduces GuitarDuets, roughly three hours of real and synthesized classical guitar duet recordings, with note-level annotations for the synthesized portion. Using an adapted Demucs separation architecture, it benchmarks monotimbral separation and a joint transcription-and-separation pipeline. Its central finding is that training on both the real and synthesized subsets improves separation on a bleeding-free real test set compared to training on either subset alone. The paper also reports that ground-truth note labels substantially help separation, whereas predicted labels yield only marginal gains.

What carries the argument

The load-bearing mechanism is the adapted Hybrid Transformer Demucs separator, a dual U-Net operating in both waveform and spectrogram domains, trained with a permutation-invariant loss plus a mixture-consistency term: $\alpha \min(|\hat{g}_1-g_1|+|\hat{g}_2-g_2|,|\hat{g}_2-g_1|+|\hat{g}_1-g_2|) + \beta |(\hat{g}_1+\hat{g}_2)-(g_1+g_2)|$, with $\alpha=0.8$ and $\beta=0.2$. This loss allows the two output channels to swap guitars without penalty and keeps their sum close to the input mixture. The dataset is the second piece of machinery: 58.6 minutes of real duets recorded with two microphones plus 106 minutes of synthesized duets generated with a virtual nylon-guitar instrument, the synthesized half carrying MIDI note annotations usable as score conditioning. Ground-truth labels are injected into both the temporal and spectral branches of Demucs, and when labels are unavailable a Residual Shuffle-Exchange transcription network generates them.

What would settle it

Run the same training setups on a larger held-out set of real guitar duets recorded with separate microphones or direct pickups; if real-only training matches or beats combined training on that set, the claimed benefit of mixing synthetic data would disappear.

Watch

Extended reading notes

Core claim

The paper's central finding is that combining real recordings of simultaneously playing classical guitarists with synthesized duets generated from MIDI scores improves monotimbral separation over using either data type alone, when evaluated on a specially recorded seven-track test set free of cross-microphone bleed. Cross-dataset experiments with the adapted Demucs model show that training on the complete GuitarDuets dataset gives the highest SDR for the first guitar, while the second guitar is separated more consistently when GuitarSet is also included in training. A second result is that conditioning the separator on ground-truth note activity labels improves separation, especially SIR, while labels predicted by a separate Residual Shuffle-Exchange transcription network help only marginally on real duets and slightly hurt on synthetic duets. Finally, the paper argues that SDR and SI-SDR behave differently for monotimbral mixtures than for multitimbral ones, so direct SDR comparisons with multitimbral benchmarks are not appropriate.

Load-bearing premise

The evaluation rests on a seven-track test set recorded without microphone bleed, and if that test set is not representative of ordinary real duet recordings, the measured gains from adding synthetic data may not transfer to realistic conditions.

Editorial extensions

If this is right

  • The GuitarDuets dataset provides a benchmark for monotimbral separation where no suitable polyphonic same-instrument dataset previously existed.
  • Synthesized data can usefully supplement scarce real recordings for same-timbre separation, at least when the real test conditions are represented in training.
  • Score-informed separation helps most when note labels are exact, so systems that rely on predicted transcriptions must close the gap between transcription and separation.
  • SDR and SI-SDR scores for monotimbral separation should not be compared directly with multitimbral separation benchmarks.
  • The joint transcription-and-separation framework gives a template for using note predictions as auxiliary information in other monotimbral tasks.

Reading between the lines

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

  • The paper leaves implicit that the synthetic subset's MIDI annotations could be used to train far more polyphonic guitar transcribers, which might turn the marginal predicted-label gains into substantial ones once the transcriber generalizes better.
  • The mixture-consistency term in the loss is a general regularization that could be tested on other monotimbral tasks such as violin duets or choir separation.
  • The consistently weaker performance on the second guitar suggests the model treats it as residual noise; this points toward source-identity embeddings, similar to speaker embeddings, as a promising direction the paper does not explore.
  • A practical extension would be listening tests to check whether the measured SDR gains from combined training correspond to perceptually cleaner separation, a question the paper flags but does not answer.
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 / 7 minor

Summary. The paper introduces GuitarDuets, a dataset of roughly 58.6 minutes of real and 106 minutes of synthesized classical guitar duet recordings, with MIDI note annotations for the synthetic portion. It adapts Hybrid Transformer Demucs to two-guitar separation and reports cross-dataset experiments on a bleeding-free held-out test set, comparing training on real-only, synthetic-only, and combined data. It also proposes score-informed conditioning with ground-truth labels and a two-stage transcription-separation pipeline with predicted labels, and analyzes SDR versus SI-SDR behavior on simulated monotimbral and multitimbral mixtures. The headline claim is that combining real and synthesized subsets improves separation on the independent test set compared with using either subset alone.

Significance. The main contribution is a publicly released dataset that fills a real gap: monotimbral source separation with polyphonic co-playing instruments. The design has several strengths: a dedicated bleeding-free test set, synthetic data with note-level annotations, use of an established strong baseline (Demucs), and a systematic cross-dataset training comparison. If the performance claims survive a more rigorous evaluation, the finding that synthetic data can complement a small real corpus in same-timbre separation would be practically useful, and the dataset could become a standard benchmark. The paper is honest about the difficulty of the task, noting the large SDR gap between the two guitars.

major comments (4)
  1. [Abstract; §4.2, Table 3] The abstract's claim that using both the real and synthesized subsets 'leads to improved separation performance' is not supported for the second guitar. In Table 3, comparing the R-only row with the R+S row, G2 SDR decreases from 1.014 to 0.920 dB, G2 SAR from 1.424 to 0.896 dB, and G2 SIR from 4.873 to 4.104 dB; G2 SI-SDR improves from -3.536 to -3.133 dB, while G1 improves on all four metrics (e.g., G1 SDR from 4.952 to 5.882 dB). Since a duet separation system must provide both sources, the headline claim should either be qualified to the first guitar or be based on a clearly defined aggregate metric; as written it is contradicted by part of the paper's own results.
  2. [§4.2, Tables 3 and 4] All separation results are reported as single numbers from a single training run, without seeds, error bars, or significance tests, and the test set consists of only 7 tracks. The key head-to-head gains (e.g., +0.93 dB G1 SDR for R+S versus R-only) are therefore not shown to be statistically reliable. Please report results over multiple random seeds, or at least per-track score distributions with paired tests such as Wilcoxon or bootstrap, for the central comparisons, and state the variance explicitly. This is necessary to support the comparison-based conclusions of the paper.
  3. [§4.4, Eq. (2), Figure 4] The metric-behavior analysis is based on a single track (Track 29 of GuitarDuets(S)) and on synthetic additive mixtures m = αx1 + (1−α)x2, yet the text concludes that 'both metrics for the guitar mixtures are consistently higher' than for multitimbral mixtures. With no error bars or a range of tracks, 'consistently' is not established. Please extend this analysis to multiple tracks and report the variability; if the illustrative example is intended only as a motivating demonstration, say so explicitly and soften the claim.
  4. [§2.2] The evaluation rests entirely on the 7-track bleeding-free test set. This is a well-motivated design, but the paper does not discuss whether those 7 tracks are representative of the real-recording conditions used in training (instruments, microphones, room, genre, tempo, or balance between guitars). A 7-track test set can support the paper's central claim only if the authors either provide evidence of representativeness or acknowledge the limitation and avoid generalizing beyond the test conditions. Please add a description of how the test tracks were selected and a discussion of this limitation.
minor comments (7)
  1. [Abstract] The phrase 'have focused in the multi-timbral case' should be 'have focused on the multi-timbral case'; similar grammatical issues appear elsewhere in the introduction.
  2. [§1, §2.2] There are minor typographical issues, including 'avalaible' in the introduction and the unusual spacing in 'W A V' in Section 2.2; these should be corrected.
  3. [§3.1, Eq. (1)] The loss weights α and β are said to be set after preliminary experiments, but no sensitivity analysis or ablation is reported; please state the range explored or provide a brief justification.
  4. [Table 3] The header layout is confusing: 'Source Datasets' and 'Metrics' are on the same row, making it easy to misassign the checkmarks; use separate header rows or subheadings to clarify which datasets are included in each condition.
  5. [Table 4] The test sets differ between the GuitarDuets(S) and GuitarDuets(R) panels, so comparisons across the two panels should be explicitly flagged as not directly comparable.
  6. [§4.2] The training/validation split is described as 80-20, but it is not explicitly stated that the bleeding-free test set is excluded from training; please make this explicit.
  7. [Figure 3] The qualitative claim that predicted note labels 'enable the separation model to more accurately sustain notes' is not supported by any quantitative measure; consider adding an objective analysis or softening the interpretation.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: separation results are evaluated on a held-out test set, synthetic data comes from external MIDI scores, and the only tuning is standard hyperparameter selection.

full rationale

The paper's derivation chain is empirical and self-contained. The real and synthetic portions of GuitarDuets are constructed from external sources (microphone recordings; MuseScore MIDI scores rendered through a virtual instrument), and all separation models are evaluated on a held-out, bleeding-free test set (Secs. 2.2, 4.2) that is not used for training or weight adjustment. The only tuned quantities, the loss weights alpha=0.8 and beta=0.2 in Eq. (1), are standard hyperparameters set after preliminary experiments, and they are not refit to the test set nor renamed as predictions. The score-informed experiments condition the separator on note labels, but the paper separately reports ground-truth-conditioned and estimated-label cases (Table 4), so the comparisons are empirical rather than definitional. The metric-behavior analysis in Sec. 4.4 is a controlled simulation and does not feed back into the separation results. Self-citations (e.g., [15] for a prior singing-voice separation method) are contextual and not load-bearing. The closest concern is that the abstract's unqualified 'improved separation performance' is contradicted for guitar 2 by Table 3 (R+S vs. R-only: G2 SDR decreases from 1.014 to 0.920 dB), but that is a correctness/statistical-support issue, not circularity. No step in the paper reduces by construction to its own inputs, so the appropriate circularity score is 0.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central empirical claims rely on the representativeness of the synthetic guitar sounds and of the bleeding-free test set, on the suitability of Demucs and RSE architectures for the task, and on the specific loss weighting chosen after preliminary experiments.

free parameters (1)
  • loss weights alpha, beta = α=0.8, β=0.2
    Set after preliminary experiments (Section 4.1); they balance permutation-invariant L1 loss against mixture-consistency loss and affect the training objective that produces all reported results.
assumptions (4)
  • domain assumption The synthetic virtual instrument reproduces guitar timbres well enough that training on it transfers to real recordings.
    The central claim that combining synthetic and real data improves separation depends on the synthetic data conveying useful timbral and musical structure (Sections 2.2 and 4.2).
  • domain assumption The bleeding-free test set is a valid ground truth for general real-world guitar duets.
    The test set contains only 7 tracks from a specialized recording setup; the paper assumes this generalizes beyond the specific microphones, room, and guitars (Section 2.2).
  • domain assumption The MIDI scores from MuseScore are reasonably accurate transcriptions of the duets.
    The synthetic subset is generated from community MIDI scores, and the note-level annotations inherit any errors in those transcriptions (Section 2.2).
  • domain assumption Demucs and RSE provide suitable inductive biases for monotimbral separation and transcription.
    The paper adapts these architectures without fundamental modification, assuming that their design choices, developed for multi-timbral and general audio tasks, transfer to same-timbre separation (Sections 3.1 and 3.2).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Classical Guitar Duet Separation using GuitarDuets -- a Dataset of Real and Synthesized Guitar Recordings." pith.science (2026). https://pith.science/paper/5HZTNI4N

@misc{pith2026250701172,
  author       = {Pith},
  title        = {Pith review of: Classical Guitar Duet Separation using GuitarDuets -- a Dataset of Real and Synthesized Guitar Recordings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5HZTNI4N}},
  note         = {Machine review of arXiv:2507.01172}
}
read the original abstract

Recent advancements in music source separation (MSS) have focused in the multi-timbral case, with existing architectures tailored for the separation of distinct instruments, overlooking thus the challenge of separating instruments with similar timbral characteristics. Addressing this gap, our work focuses on monotimbral MSS, specifically within the context of classical guitar duets. To this end, we introduce the GuitarDuets dataset, featuring a combined total of approximately three hours of real and synthesized classical guitar duet recordings, as well as note-level annotations of the synthesized duets. We perform an extensive cross-dataset evaluation by adapting Demucs, a state-of-the-art MSS architecture, to monotimbral source separation. Furthermore, we develop a joint permutation-invariant transcription and separation framework, to exploit note event predictions as auxiliary information. Our results indicate that utilizing both the real and synthesized subsets of GuitarDuets leads to improved separation performance in an independently recorded test set compared to utilizing solely one subset. We also find that while the availability of ground-truth note labels greatly helps the performance of the separation network, the predicted note estimates result only in marginal improvement. Finally, we discuss the behavior of commonly utilized metrics, such as SDR and SI-SDR, in the context of monotimbral MSS.

Figures

Figures reproduced from arXiv: 2507.01172 by the authors.

Figure 1
Figure 1. In particular, in the temporal branch, the activ [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 54 canonical work pages

  1. [1]

    Classical Guitar Duet Separation using GuitarDuets -- a Dataset of Real and Synthesized Guitar Recordings

    INTRODUCTION The task of music source separation (MSS) involves dis- secting a musical composition into its constituent sources, typically segregating individual instruments or vocal tracks from a composite audio mixture [1–3]. Due to the mul- titude of the co-playing sources, as well as its utility in a variety of applications [1], MSS stands as a signif...

    work page Pith review arXiv 2024

  2. [2]

    Session Guitarist - Picked Nylon

    DATASETS 2.1 Existing Datasets Datasets available for music source separation or transcrip- tion are primarily divided into multitimbral and mono- timbral ones, each offering instrument-specific tracks or stems, often accompanied by transcriptions. Multitim- bral datasets such as musdb18 [7], URMP [27], Med- leyDB [38], MoisesDB [26] and SLAKH [28] are mo...

    work page

  3. [3]

    Overview of the incorporation of note-level an- notations into the Demucs Architecture

    METHODOLOGY 3.1 Separation Architecture In this work the Hybrid Transformer Demucs [5] was used as the separation backbone, consisting of dual U-Nets [16], operating in both time and spectrogram domains, each 2 https://www.native-instruments.com/en/products/komplete/guitar/ session-guitarist-picked-nylon/ 3 https://musescore.com/ 4 https://www.apple.com/l...

    work page

  4. [4]

    We adapted the backbone Demucs model for classical guitar duet separation, modifying it to out- put two stereo signals, one for each guitar

    EXPERIMENTAL EV ALUATION 4.1 Experimental Setup For the separation experiments, we used both the real and synthesized subsets of GuitarDuets, which we will further denote as GuitarDuets(R) and GuitarDuets(S), re- spectively, as well as the GuitarSet, for which mixtures were generated via addition of the available comping and solo excerpts. We adapted the ...

    work page

  5. [5]

    3rd Call for H.F.R.I. Research Projects to support Post-Doctoral Researchers

    CONCLUSIONS In this paper, we introduced GuitarDuets, a dataset con- sisting of both real and synthesized classical guitar duets. We exhibit that our dataset can be utilized for developing monotimbral source separation algorithms within both tra- ditional and score-informed frameworks. We further de- veloped a joint permutation-invariant framework for tra...

    work page

  6. [6]

    Open-unmix-a reference implementation for music source separation,

    F.-R. Stöter, S. Uhlich, A. Liutkus, and Y . Mitsufuji, “Open-unmix-a reference implementation for music source separation,” Journal of Open Source Software , vol. 4, no. 41, p. 1667, 2019

    work page 2019

  7. [7]

    A review on music source separa- tion,

    R. Liu and S. Li, “A review on music source separa- tion,” in Proc. IEEE Youth Conf. on Information, Com- puting and Telecommunication, 2009

    work page 2009

  8. [8]

    Music demixing challenge 2021,

    Y . Mitsufuji, G. Fabbro, S. Uhlich, F.-R. Stöter, A. De- fossez, M. Kim, W. Choi, C. Y . Yu, and K. W. Cheuk, “Music demixing challenge 2021,” Frontiers in Signal Processing, vol. 1, Jan., 2022

    work page 2021

Show all 55 references
  1. [9]

    Musical source separation: An intro- duction,

    E. Cano, D. FitzGerald, A. Liutkus, M. D. Plumbley, and F.-R. Stöter, “Musical source separation: An intro- duction,” IEEE Signal Processing Magazine , vol. 36, no. 1, pp. 31–40, 2019

    2019

  2. [10]

    An overview of lead and accompaniment separation in music,

    Z. Rafii, A. Liutkus, F.-R. Stöter, S. I. Mimilakis, D. FitzGerald, and B. Pardo, “An overview of lead and accompaniment separation in music,” IEEE/ACM Transactions on Audio, Speech, and Language Pro- cessing, vol. 26, no. 8, pp. 1307–1335, 2018

    2018

  3. [11]

    Hybrid trans- formers for music source separation,

    S. Rouard, F. Massa, and A. Défossez, “Hybrid trans- formers for music source separation,” in Proc. Int’l Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2023

    2023

  4. [12]

    Wave-u-net: A multi-scale neural network for end-to-end audio source separation,

    D. Stoller, S. Ewert, and S. Dixon, “Wave-u-net: A multi-scale neural network for end-to-end audio source separation,” in Proc. Int’l Society for Music Informa- tion Retrieval Conf. (ISMIR), 2018

    2018

  5. [13]

    Musdb18-a corpus for music separation,

    Z. Rafii, A. Liutkus, F.-R. Stöter, S. I. Mimilakis, and R. Bittner, “Musdb18-a corpus for music separation,” 2017

    2017

  6. [14]

    Vincent, T

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

    2018

  7. [15]

    Main instrument separation from stereo- phonic audio signals using a source/filter model,

    J.-L. Durrieu, A. Ozerov, C. Févotte, G. Richard, and B. David, “Main instrument separation from stereo- phonic audio signals using a source/filter model,” in Proc. 17th European Signal Processing Conf., 2009

    2009

  8. [16]

    Music source separation in the waveform domain,

    A. Défossez, N. Usunier, L. Bottou, and F. R. Bach, “Music source separation in the waveform domain,”

  9. [17]

    Meta- learning extractors for music source separation,

    D. Samuel, A. Ganeshan, and J. Naradowsky, “Meta- learning extractors for music source separation,” in Proc. Int’l Conf. on Acoustics, Speech and Signal Pro- cessing (ICASSP), 2020

    2020

  10. [18]

    Music source separation with band- split RNN,

    Y . Luo and J. Yu, “Music source separation with band- split RNN,” IEEE/ACM Trans. on Audio, Speech, and Language Processing, vol. 31, pp. 1893–1901, 2023

    1901

  11. [19]

    Deep learning based source separation ap- plied to choir ensembles,

    D. Petermann, P. Chandna, H. Cuesta, J. Bonada, and E. Gómez, “Deep learning based source separation ap- plied to choir ensembles,”Proc. Int’l Society for Music Information Retrieval Conf. (ISMIR), 2020

    2020

  12. [20]

    Transfer learning with jukebox for music source sep- aration,

    W. Zai El Amri, O. Tautz, H. Ritter, and A. Melnik, “Transfer learning with jukebox for music source sep- aration,” in Proc. Int’l Conf. on Artificial Intelligence Applications and Innovation, 2022

    2022

  13. [21]

    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 and Lang. Proc., vol. 27, no. 8, p. 1256–1266, Aug. 2019

    2019

  14. [22]

    HTMD- Net: A Hybrid Masking-Denoising Approach to Time- Domain Monaural Singing V oice Separation,

    C. Garoufis, A. Zlatintsi, and P. Maragos, “HTMD- Net: A Hybrid Masking-Denoising Approach to Time- Domain Monaural Singing V oice Separation,” inProc. 29th European Signal Processing Conf. (EUSIPCO) , 2021

    2021

  15. [23]

    U-Net: Con- volutional networks for biomedical image segmenta- tion,

    O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Con- volutional networks for biomedical image segmenta- tion,” in Medical Image Computing and Computer- Assisted Intervention (MICCAI), 2015

    2015

  16. [24]

    Jointist: Simultaneous im- provement of multi-instrument transcription and music source separation via joint training,

    K. W. Cheuk et al. , “Jointist: Simultaneous im- provement of multi-instrument transcription and music source separation via joint training,” 2023. [Online]. Available: https://arxiv.org/pdf/2206.10805

    2023 arXiv

  17. [25]

    End- to-end sound source separation conditioned on instru- ment labels,

    O. Slizovskaia, L. Kim, G. Haro, and E. Gomez, “End- to-end sound source separation conditioned on instru- ment labels,” in Proc. Int’l Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2019

    2019

  18. [26]

    Moisesdb: A dataset for source separation beyond 4- stems,

    I. Pereira, F. Araújo, F. Korzeniowski, and R. V ogl, “Moisesdb: A dataset for source separation beyond 4- stems,” in Proc. Int’l Conf. of International Society for Music Information Retrieval (ISMIR), 2023

    2023

  19. [27]

    Improved separation of polyphonic chamber music signals by integrating instrument activity labels,

    M. Schwabe and M. Heizmann, “Improved separation of polyphonic chamber music signals by integrating instrument activity labels,” IEEE Access, vol. 11, pp. 42 999–43 007, 2023

    2023

  20. [28]

    Monaural score- informed source separation for classical music using convolutional neural networks,

    M. Miron, J. Janer, and E. Gómez, “Monaural score- informed source separation for classical music using convolutional neural networks,” in Proc. Int’l Society for Music Information Retrieval Conf. (ISMIR), 2017

    2017

  21. [29]

    Score-informed source sep- aration of choral music,

    M. Gover and P. Depalle, “Score-informed source sep- aration of choral music,” in Proc. Int’l Society for Mu- sic Information Retrieval Conf. (ISMIR), 2020

    2020

  22. [30]

    A unified model for zero-shot music source separation, transcrip- tion and synthesis,

    L. Lin, Q. Kong, J. Jiang, and G. G. Xia, “A unified model for zero-shot music source separation, transcrip- tion and synthesis,” in Proc. Int’l Society for Music In- formation Retrieval Conf. (ISMIR), 2021

    2021

  23. [31]

    Some experiments on the recognition of speech, with one and with two ears,

    E. C. Cherry, “Some experiments on the recognition of speech, with one and with two ears,” Journal of the Acoustical Society of America , vol. 25, pp. 975–979, 1953

    1953

  24. [32]

    Train from scratch: Single-stage joint training of speech separa- tion and recognition,

    J. Shi, X. Chang, S. Watanabe, and B. Xu, “Train from scratch: Single-stage joint training of speech separa- tion and recognition,” Computer Speech & Language, vol. 76, p. 101387, Apr., 2022

    2022

  25. [33]

    Multi-microphone neural speech separation for far- field multi-talker speech recognition,

    T. Yoshioka, H. Erdogan, Z. Chen, and F. Alleva, “Multi-microphone neural speech separation for far- field multi-talker speech recognition,” in Proc. Int’l Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2018

    2018

  26. [34]

    Creating a multitrack classical music per- formance dataset for multimodal music analysis: Chal- lenges, insights, and applications,

    B. Li et al., “Creating a multitrack classical music per- formance dataset for multimodal music analysis: Chal- lenges, insights, and applications,” IEEE Trans. on Multimedia, vol. 21, pp. 522–535, 2018

    2018

  27. [35]

    Cutting music source separation some slakh: A dataset to study the impact of training data quality and quantity,

    E. Manilow et al. , “Cutting music source separation some slakh: A dataset to study the impact of training data quality and quantity,” in Proc. IEEE Workshop on Applications of Signal Processing to Audio and Acous- tics (WASPAA), 2019

    2019

  28. [36]

    Ensembleset: A new high-quality synthesised dataset for chamber ensemble separation,

    S. Sarkar, E. Benetos, and M. Sandler, “Ensembleset: A new high-quality synthesised dataset for chamber ensemble separation,” in Proc. Int’l Society for Music Information Retrieval Conf. (ISMIR), 2022

    2022

  29. [37]

    Guitarset: A dataset for guitar transcription,

    Q. Xi, R. M. Bittner, J. Pauwels, X. Ye, and J. P. Bello, “Guitarset: A dataset for guitar transcription,” in Proc. Int’l Society for Music Information Retrieval Conf. (IS- MIR), 2018

    2018

  30. [38]

    Medleydb: A multitrack dataset for annotation-intensive mir research,

    R. Bittner, J. Salamon, M. Tierney, M. Mauch, C. Can- nam, and J. Bello, “Medleydb: A multitrack dataset for annotation-intensive mir research,” in Proc. Int’l So- ciety for Music Information Retrieval Conf. (ISMIR) , 2014

    2014

  31. [39]

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

    D. Yu, M. Kolbæk, Z.-H. Tan, and J. Jensen, “Permu- tation invariant training of deep models for speaker- independent multi-talker speech separation,” in Proc. Int’l Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2017

    2017

  32. [40]

    Leveraging real electric guitar tones and effects to improve robustness in guitar tablature transcription modeling,

    H. Pedroza, W. Abreu, R. Corey, and I. Roman, “Leveraging real electric guitar tones and effects to improve robustness in guitar tablature transcription modeling,” 2024. [Online]. Available: https://arxiv. org/abs/2405.14679

    2024 arXiv

  33. [41]

    V ocal har- mony separation using time-domain neural networks,

    S. Sarkar, E. Benetos, and M. B. Sandler, “V ocal har- mony separation using time-domain neural networks,” in Proc. Interspeech Conf., 2021

    2021

  34. [42]

    Medleyvox: An evaluation dataset for mul- tiple singing voices separation,

    C.-B. Jeon, H. Moon, K. Choi, B. S. Chon, and K. Lee, “Medleyvox: An evaluation dataset for mul- tiple singing voices separation,” in Proc. Int’l Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2023

    2023

  35. [43]

    Automatic transcription of a cappella recordings from multiple singers,

    R. Schramm and E. Benetos, “Automatic transcription of a cappella recordings from multiple singers,” inAES International Conference Semantic Audio, 2017

    2017

  36. [44]

    For further details about the architecture we refer to [37, 44]

    by employing both Switch and Shuffle layers to cap- ture sequence dependencies effectively, as well as reducing its computational overhead by incorporating strided convo- lutions. For further details about the architecture we refer to [37, 44]. In our implementation, the RSE’s...

  37. [45]

    Residual shuffle-exchange networks for fast processing of long sequences,

    A. Draguns, E. Ozolins, A. Sostaks, M. Apinis, and K. Freivalds, “Residual shuffle-exchange networks for fast processing of long sequences,” inProc. AAAI Conf. on Artificial Intelligence, 2020

    2020

  38. [46]

    Syn- thtab: Leveraging synthesized data for guitar tabla- ture transcription,

    Y . Zang, Y . Zhong, F. Cwitkowitz, and Z. Duan, “Syn- thtab: Leveraging synthesized data for guitar tabla- ture transcription,” in Proc. Int’l Conf. on Acoustics, Speech and Signal Processing (ICASSP) , 2024, pp. 1286–1290

    2024

  39. [47]

    Film: Visual reasoning with a general conditioning layer,

    E. Perez, F. Strub, H. De Vries, V . Dumoulin, and A. Courville, “Film: Visual reasoning with a general conditioning layer,” in Proc. AAAI Conf. on Artificial Intelligence, 2018

    2018

  40. [48]

    Hybrid spectrogram and waveform source separation,

    A. Défossez, “Hybrid spectrogram and waveform source separation,” inProc. Music Demixing Workshop (MDX), 2021

    2021

  41. [49]

    Going deeper with image Transform- ers,

    H. Touvron, M. Cord, A. Sablayrolles, G. Synnaeve, and H. Jégou, “Going deeper with image Transform- ers,” in Proc. Int’l Conf. on Computer Vision (ICCV) , 2021

    2021

  42. [50]

    Learning features of music from scratch,

    J. Thickstun, Z. Harchaoui, and S. Kakade, “Learning features of music from scratch,” in Proc. Int’l Conf. on Learning Representations (ICLR), 2017

    2017

  43. [51]

    Neural Shuffle-Exchange Networks - Sequence processing in O(n log n) time,

    K. Freivalds, E. Ozolins, and A. Sostaks, “Neural Shuffle-Exchange Networks - Sequence processing in O(n log n) time,” in Advances in Neural Information Proc. Systems (NeurIPS), 2019

    2019

  44. [52]

    Perfor- mance measurement in blind audio source separation,

    E. Vincent, R. Gribonval, and C. Févotte, “Perfor- mance measurement in blind audio source separation,” IEEE Transactions on Audio, Speech, and Language Processing, vol. 14, no. 4, pp. 1462–1469, 2006

    2006

  45. [53]

    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. Int’l Conf. on Acoustics Speech and Signal Processing (ICASSP), 2019

    2019

  46. [55]

    Conditioned-u- net: Introducing a control mechanism in the u-net for multiple source separations,

    G. Meseguer-Brocal and G. Peeters, “Conditioned-u- net: Introducing a control mechanism in the u-net for multiple source separations,” in Proc. Int’l Society for Music Information Retrieval Conf. (ISMIR), 2019

    2019

  47. [2019]

    Available: http://arxiv.org/abs/1911

    [Online]. Available: http://arxiv.org/abs/1911. 13254

    1911

Pith tools

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