Pith. sign in

REVIEW 2 major objections 5 minor 46 references

Meta-learning-based percussion transcription and $t\bar{a}la$ identification from low-resource audio

T0 review · 2 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Meta-learning with a frozen pre-trained front end makes tabla stroke transcription work from minutes of labelled audio, beating supervised and transfer baselines.

desk verdict A useful MAML-for-percussion paper that needs a corrected transfer-learning baseline before its main claim can be trusted. read the letter →

arxiv 2501.04742 v3 pith:IX644BTR submitted 2025-01-08 eess.AS

Rahul Bapusaheb Kodag , Vipul Arora This is my paper · ORCID
classification eess.AS
keywords tablastroketranscriptionmodel-agnosticmeta-learningtalaidentificationlow-resourceaudioautomaticdrumHindustaniclassicalmusicneedleman-wunschalignmentCRNN
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that a meta-learning procedure can break the data bottleneck in percussion transcription: instead of needing large annotated corpora for every instrument or recording style, a model can be prepared so that a few minutes of new labelled audio adapts it to unseen stroke classes. The authors build a CRNN stroke transcriber, pre-train it on synthetic mridangam strokes, meta-train it with model-agnostic meta-learning (MAML) on ten tabla stroke classes, and then adapt it to new tabla datasets. On tabla solo recordings, a four-way tabla dataset, and full concert mixes, the meta-learned model reaches 81.32, 83.10, and 63.02 F1 respectively, in each case above both transfer learning and supervised training on the same limited data. The same recipe transfers to Western drum transcription, and the resulting stroke sequences support two new tala identification scores, including a Needleman-Wunsch alignment score that outperforms the rough-LCS baseline on concert audio. If the claim holds, low-resource transcription of Indian percussion, and the rhythmic analysis built on it, becomes practical without hand-harmonizing label schemes across traditions.

What carries the argument

The load-bearing object is a CRNN stroke transcriber trained in two stages. In the first stage, convolutional layers are pre-trained on a synthetic mridangam stroke dataset and then frozen, on the assumption that they capture low-level spectral features shared by mridangam and tabla; in the second, the recurrent and classifier layers are meta-trained with MAML, an optimization strategy that prepares the model to learn new tasks in a few gradient steps, using support and query sets so that the final parameter initialization adapts quickly to new stroke classes. For tala identification, the machinery is the Needleman-Wunsch alignment score, which preserves both the order and the continuity of the transcribed stroke sequence, and the Stroke Ratio Score, a cosine similarity between the stroke-count ratios of the test audio and the fixed ratio vectors of four common talas.

What would settle it

Train the same CRNN with the convolutional layers unfrozen during meta-learning and compare F1 on the four-way tabla dataset: if the frozen front end is not load-bearing, the gap between meta-learning and transfer learning should vanish. A sharper test is to take a tabla recording on a very different tonic, or a different membrane drum, and see whether the frozen mridangam front end still beats fine-tuning the whole network; if it does not, the shared low-level features assumption fails.

Watch

Extended reading notes

Core claim

The paper's central discovery is that stroke transcription can be treated as a meta-learning problem in which the task is a dataset-specific stroke vocabulary. The authors show that a CRNN whose convolutional layers are pre-trained on synthetic mridangam audio and frozen, while its recurrent and classifier layers are trained with MAML, yields a parameter initialization that adapts to new tabla stroke classes with only minutes of labelled audio. Across the three tabla test scenarios this initialization beats the same architecture trained from scratch on the limited data and fine-tuned via transfer learning, and it also beats four-way tabla transcription baselines built from Western-drum transfer. On Western drums the same method outperforms a state-of-the-art CRNN, a partially fixed NMF method, and transfer learning across solo-drum, drum-with-percussion, and full-mix datasets. The authors additionally claim that the transcribed stroke sequences can be matched to thekas of Tintal, Ekatal, Jhaptal, and Rupak by Needleman-Wunsch alignment or by stroke-ratio cosine similarity, with the alignment score reaching 48.96 percent accuracy on concert recordings, substantially above the 33.6 percent of the rough-LCS baseline.

Load-bearing premise

The argument collapses if the low-level spectral features learned on synthetic mridangam audio are not transferable to tabla, since the paper freezes those convolutional layers during meta-learning.

Editorial extensions

If this is right

  • A new tabla dataset with its own stroke vocabulary can be transcribed after a few minutes of labelled audio, without retraining on a large corpus or reconciling stroke labels with existing datasets.
  • The same meta-learned initialization transfers to Western drum transcription, matching or beating source-separated baselines even when no pre-trained drum model exists.
  • Tala identification becomes a downstream sequence problem: the transcribed stroke stream, not hand-crafted features, is enough to score candidate talas, with the alignment score more accurate and the ratio score much faster.
  • Polyphonic concert audio degrades but does not break the approach: the largest performance drop comes when vocals and instruments overlap the strokes, pointing to improved onset detection as the main lever.
  • Meta-learning sidesteps label heterogeneity, so datasets from different gharana conventions or drum ontologies do not need to be mapped onto a common label set before training.

Reading between the lines

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

  • If the frozen-features assumption generalizes, the same mridangam pre-training could transfer to other membrane percussion of the Indian subcontinent, such as pakhawaj or dholak, providing a cheap few-shot transcription starting point.
  • The tala identification results imply a bound: with transcription F1 around 63 percent on concert audio, sequence alignment can only partially recover the rhythmic cycle, so better stroke transcription or cycle-level decoding should push identification well above the reported 48.96 percent.
  • The ADT results suggest that meta-learning may reduce the need for source separation in polyphonic event detection generally, a pattern that could be tested on other low-resource audio tasks with heterogeneous label ontologies, such as bioacoustic monitoring.
  • The two proposed scores could be combined, using the fast ratio score to shortlist talas and the slower alignment score to confirm, an ensemble the paper does not evaluate.
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

2 major / 5 minor

Summary. This paper proposes a model-agnostic meta-learning (MAML) approach for low-resource tabla stroke transcription (TST) and tala identification in Hindustani classical music, and extends it to automatic drum transcription (ADT). The TST model is a CRNN pre-trained on a synthetic mridangam dataset and then meta-trained on 12 minutes of tabla solo data (DT1) before few-shot adaptation on target datasets (DT1, DT2, DT3). Two tala identification methods are introduced: a Needleman-Wunsch sequence matching score and a stroke-ratio cosine similarity score. Experiments report F1 scores for TST and ADT over several datasets, with the best TST result being 81.32% F1 on DT1, and best ADT result 80.1% on DTD-DD1, outperforming the presented supervised, transfer-learning, and NMF baselines. The authors also provide a public repository link for dataset and supplementary materials.

Significance. If the reported gains are robust, the paper provides a practical recipe for few-shot percussion transcription under low-resource conditions and proposes two simple, computable tala identification scores that go beyond existing symbolic matching. The use of standard public datasets (DT1-DT3, ADTOF, IDMT-SMT, ENST, MDB) and the public release of the curated mridangam dataset are strengths. However, the central comparison with transfer learning is confounded by the amount of training data, and the lack of error bars or statistical tests leaves the headline claims without uncertainty quantification; these issues must be addressed before the empirical contribution can be accepted.

major comments (2)
  1. [V-A2a, Tables IV-VI, VIII-XI] The claim that MAML outperforms transfer learning is not established because the transfer-learning baseline does not see the meta-training data. For TST, PTM1+ML is meta-trained on 12 minutes of DT1 (Table I) and then adapted on the 5-min/10-min support set, while PTM1+TL is described as 'fine-tuned and tested using the same adaptation and evaluation data,' i.e., only the support set. The ADT comparison has the same asymmetry: CRNN+ML meta-trains on the entire 114-hour ADTOF set (Table II), whereas CRNN+TL is fine-tuned only on the 5-min adaptation set. The reported gains (e.g., 81.32 vs. 62.17 on DT1 in Table IV; 80.1 vs. 62.3 on DD1 in Table VIII) could therefore be explained by the extra target-domain data used in meta-training. Please add a transfer-learning control that is trained on the same meta-training data with an ordinary supervised loss and then fine-tuned on the same support set, and report that comparison.
  2. [VI-A, Tables IV-XI] All main results are reported as single F1 numbers without error bars, number of repetitions/seeds, or statistical significance tests. In low-resource settings with 5-minute adaptation sets, run-to-run variability is likely substantial, and a few-point differences between methods may not be significant. Report the mean and standard deviation over at least several runs with different random seeds and split samples, or provide an appropriate significance test, and describe the exact train/validation/test split procedure used for each dataset.
minor comments (5)
  1. [III-C] There is a contradictory sentence: 'where convolutional layer parameters are frozen during meta-training and meta-testing, all parameters θ1, θ2, θ3 are trainable in ADT.' The preceding text says the ADT model is trained from scratch with all parameters trainable; please clarify which statement is correct and ensure Algorithm 1's description matches.
  2. [V-A1b] The 3% of maximum amplitude threshold used to define the 'No-stroke' region is not fully specified; explain how it is applied (per audio file, per stroke, fixed across the dataset) and whether it was tuned.
  3. [Table VII] The tala identification accuracies (48.96% for NW, 43.1% for ratio) are well above chance (25%) but still modest; report confidence intervals or a significance test against chance, and consider comparing with a simple dynamic time warping baseline to contextualize the gains.
  4. [Fig. 2] There is a typo: 'transcripted' should be 'transcribed'.
  5. [Eq. (5) / Algorithm 2] The notation σ_nw0, σ_nw1, ..., σ_nwk is not defined in the text; please define each term explicitly in relation to the sliding window over the n-stroke test sequence.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation found; the MAML and tala-identification claims are not equivalent to their inputs, although the MAML-vs-transfer-learning comparison is confounded by unequal training data.

full rationale

The paper's central TST claim is that MAML adaptation improves low-resource transcription. The derivation is not circular: the MAML objective (Eqs. 3-4) is an inner-loop support loss followed by an outer-loop query loss computed on ground-truth labels; meta-test adaptation in Sec. III-A2b uses a support set disjoint from the query/evaluation set, and the reported F1 values are obtained on the query set. The tala-identification methods are also externally anchored: NW Matching Score (Algorithm 2) and Stroke Ratio Score (Eq. 6) compare the independently transcribed test stroke sequence to hand-specified theka reference sequences/ratios from music theory (e.g., Teental [3,3,1,1]); no reference pattern is fitted from the test labels, so the match scores are not forced by construction. The mridangam pre-training is standard transfer learning, and the only in-paper self-citation, [33] (Saxena and Arora), is background on MAML for melody extraction and is not load-bearing. One caveat belongs to experimental correctness rather than circularity: Sec. V-A2a/3 describe the transfer-learning baseline as fine-tuned on 'the same adaptation and evaluation data,' while MAML additionally meta-trains on 12 min of DT1 (Table I) and, for ADT, on the 114-h ADTOF set (Table II); therefore the reported MAML-vs-TL gaps may reflect extra training data rather than the meta-learning objective. This is an unequal-data confound, not a definitional reduction, and does not make the paper's predictions equivalent to its inputs.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

No new physical or conceptual entities are introduced. The central claim rests on two domain assumptions: transfer of pre-trained features from mridangam to tabla, and the correctness of hand-specified theka patterns. Two hand-set constants (3% threshold, NW scoring) are free parameters that affect the reported numbers, though the main MAML gain does not depend on them.

free parameters (2)
  • PTM1 no-stroke threshold = 3% of stroke maximum amplitude
    Hand-set in Section V-A1b to separate stroke frames from the release region; changes the label targets and materially affects PTM1 results.
  • NW alignment scoring constants = match +1, mismatch -1, gap -2
    Chosen constants in Algorithm 2 with no sensitivity analysis; they determine the tala matching scores.
assumptions (3)
  • domain assumption Mridangam convolutional features transfer to tabla strokes
    Freezing theta_1 during MAML (Section III-A2a) is only helpful if low-level features learned on mridangam are shared with tabla.
  • domain assumption Reference theka patterns for the four talas are correct and complete
    Section III-B defines Teental [3,3,1,1], Ekta [3,1,2,1,1,2,2], Jhaptal [5,4,1], and Rupak [2,3,2] by hand; NW and ratio scores assume these patterns distinguish talas.
  • domain assumption Annotated onset times in the public datasets are accurate
    Evaluation uses F1 with a 50 ms collar against these annotations (Section VI); the paper itself notes annotation inconsistencies may affect PTM2 (Section VI-A).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Meta-learning-based percussion transcription and $t\bar{a}la$ identification from low-resource audio." pith.science (2026). https://pith.science/paper/IX644BTR

@misc{pith2026250104742,
  author       = {Pith},
  title        = {Pith review of: Meta-learning-based percussion transcription and $t\barala$ identification from low-resource audio},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IX644BTR}},
  note         = {Machine review of arXiv:2501.04742}
}
abstract

This study introduces a meta-learning-based approach for low-resource Tabla Stroke Transcription (TST) and $t\bar{a}la$ identification in Hindustani classical music. Using Model-Agnostic Meta-Learning (MAML), we address the challenges of limited annotated datasets and label heterogeneity, enabling rapid adaptation to new tasks with minimal data. The method is validated across various datasets, including tabla solo and concert recordings, demonstrating robustness in polyphonic audio scenarios. We propose two novel $t\bar{a}la$ identification techniques based on stroke sequences and rhythmic patterns. Additionally, the approach proves effective for Automatic Drum Transcription (ADT), showcasing its flexibility for Indian and Western percussion music. Experimental results show that the proposed method outperforms existing techniques in low-resource settings, significantly contributing to music transcription and studying musical traditions through computational tools.

Figures

Figures reproduced from arXiv: 2501.04742 by the authors.

Figure 1
Figure 1. System overview: (A) TST: Pre-trained on mridangam data using OTM, PTM1, PTM2, and adapted via MAML. (B) Tala ¯ identification: Identifies tala ¯ using transcribed strokes from TST and NW Matching/Stroke Ratio scores. (C) ADT: Applies the same meta-learning strategies to Western drum audio. strokes. The input Xi ∈ R 3×F ×T is a 3-channel chunk of 15 time frames (T) from log-Mel spectrograms computed using window siz… view at source ↗
Figure 2
Figure 2. The NW Matching Score compares a transcripted [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. General CNN [7] and CRNN model architecture for [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 43 canonical work pages

  1. [1]

    A review of automatic drum transcription,

    C.-W. Wu, C. Dittmar, C. Southall, R. V ogl, G. Widmer, J. Hockman, M. M ¨uller, and A. Lerch, “A review of automatic drum transcription,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 26, no. 9, pp. 1457–1483, 2018

    work page 2018

  2. [2]

    Jointist: Joint learning for multi-instrument transcription and its applications,

    K. W. Cheuk, K. Choi, Q. Kong, B. Li, M. Won, A. Hung, J.-C. Wang, and D. Herremans, “Jointist: Joint learning for multi-instrument transcription and its applications,”arXiv preprint arXiv:2206.10805, 2022. 10

    arXiv 2022

  3. [3]

    Unaligned supervision for automatic music transcription in the wild,

    B. Maman and A. H. Bermano, “Unaligned supervision for automatic music transcription in the wild,” inICML. PMLR, 2022, pp. 14 918– 14 934

    work page 2022

  4. [5]

    Acoustic features for determining goodness of tabla strokes,

    K. Narang and P. Rao, “Acoustic features for determining goodness of tabla strokes,” inISMIR, 2017, pp. 257–263

    work page 2017

  5. [6]

    Four-way classification of tabla strokes with models adapted from automatic drum transcription,

    R. MA, A. Bhattacharjee, and P. Rao, “Four-way classification of tabla strokes with models adapted from automatic drum transcription,” in ISMIR, 2021

    work page 2021

  6. [7]

    Four-way classification of tabla strokes with transfer learning using western drums,

    R. M. Ananthanarayana, A. Bhattacharjee, and P. Rao, “Four-way classification of tabla strokes with transfer learning using western drums,”Transactions of the International Society for Music Information Retrieval, vol. 6, no. 1, 2023

    work page 2023

  7. [8]

    Model-agnostic meta-learning for fast adaptation of deep networks,

    C. Finn, P. Abbeel, and S. Levine, “Model-agnostic meta-learning for fast adaptation of deep networks,” inICML. PMLR, 2017, pp. 1126– 1135

    work page 2017

  8. [9]

    Sequence labeling with meta-learning,

    J. Li, P. Han, X. Ren, J. Hu, L. Chen, and S. Shang, “Sequence labeling with meta-learning,”IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 3, pp. 3072–3086, 2021

    work page 2021

Show all 46 references
  1. [10]

    Rapid learning or feature reuse? towards understanding the effectiveness of maml,

    A. Raghu, M. Raghu, S. Bengio, and O. Vinyals, “Rapid learning or feature reuse? towards understanding the effectiveness of maml,”arXiv preprint arXiv:1909.09157, 2019

    1909 arXiv

  2. [11]

    Meta- lmtc: meta-learning for large-scale multi-label text classification,

    R. Wang, X. Su, S. Long, X. Dai, S. Huang, and J. Chen, “Meta- lmtc: meta-learning for large-scale multi-label text classification,” in Proceedings of the 2021 conference on empirical methods in natural language processing, 2021, pp. 8633–8646

    2021

  3. [12]

    Music matching based on rough longest common subsequence

    H.-J. Lin, H.-H. Wu, and C.-W. Wang, “Music matching based on rough longest common subsequence.”J. Inf. Sci. Eng., vol. 27, no. 1, pp. 95– 110, 2011

    2011

  4. [14]

    Improved musical onset detection with convolutional neural networks,

    J. Schl ¨uter and S. B ¨ock, “Improved musical onset detection with convolutional neural networks,” inICASSP. IEEE, 2014, pp. 6979– 6983

    2014

  5. [15]

    Drum transcription from polyphonic music with recurrent neural networks,

    R. V ogl, M. Dorfer, and P. Knees, “Drum transcription from polyphonic music with recurrent neural networks,” inICASSP. IEEE, 2017, pp. 201–205

    2017

  6. [16]

    Drum transcription via joint beat and drum modelling using convolutional recurrent neural networks

    R. V ogl, M. Dorfer, G. Widmer, and P. Knees, “Drum transcription via joint beat and drum modelling using convolutional recurrent neural networks.” inISMIR, 2017, pp. 150–157

    2017

  7. [17]

    Towards multi-instrument drum transcription,

    R. V ogl, G. Widmer, and P. Knees, “Towards multi-instrument drum transcription,” inProceedings of the 21st International Conference on Digital Audio Effects (DAFx-18), 2018, pp. 57–64

    2018

  8. [18]

    In-depth performance anal- ysis of the state-of-the-art algorithm for automatic drum transcription,

    M. Zehren, M. Alunno, and P. Bientinesi, “In-depth performance anal- ysis of the state-of-the-art algorithm for automatic drum transcription,” inISMIR, 2024

    2024

  9. [19]

    Analyzing and reducing the synthetic-to-real transfer gap in music information retrieval: the task of automatic drum transcription,

    ——, “Analyzing and reducing the synthetic-to-real transfer gap in music information retrieval: the task of automatic drum transcription,” arXiv preprint arXiv:2407.19823, 2024

    2024 arXiv

  10. [20]

    Detecting tala compu- tationally in polyphonic context-a novel approach,

    S. Bhaduri, A. Bhaduri, and D. Ghosh, “Detecting tala compu- tationally in polyphonic context-a novel approach,”arXiv preprint arXiv:1611.05182, 2016

    2016 arXiv

  11. [21]

    Automatic labelling of tabla signals,

    O. Gillet and G. Richard, “Automatic labelling of tabla signals,” 2003

    2003

  12. [22]

    Segmentation and recognition of tabla strokes

    P. Chordia, “Segmentation and recognition of tabla strokes.” inISMIR, vol. 20056, 2005, pp. 107–114

    2005

  13. [23]

    Tabla gyan: A system for realtime tabla recognition and resynthesis,

    P. Chordia and A. Rae, “Tabla gyan: A system for realtime tabla recognition and resynthesis,” inICMC, 2008

    2008

  14. [24]

    Acoustic features for determining goodness of tabla strokes

    K. Narang and P. Rao, “Acoustic features for determining goodness of tabla strokes.” inISMIR, 2017, pp. 257–263

    2017

  15. [25]

    Discovery of syllabic percussion patterns in tabla solo recordings,

    S. Gupta, A. Srinivasamurthy, M. Kumar, H. A. Murthy, and X. Serra, “Discovery of syllabic percussion patterns in tabla solo recordings,” in ISMIR, 2015

    2015

  16. [26]

    Few-shot acoustic event detection via meta learning,

    B. Shi, M. Sun, K. C. Puvvada, C.-C. Kao, S. Matsoukas, and C. Wang, “Few-shot acoustic event detection via meta learning,” inICASSP. IEEE, 2020, pp. 76–80

    2020

  17. [27]

    Few-shot drum transcription in polyphonic music,

    Y . Wang, J. Salamon, M. Cartwright, N. J. Bryan, and J. P. Bello, “Few-shot drum transcription in polyphonic music,”arXiv preprint arXiv:2008.02791, 2020

    2008 arXiv

  18. [28]

    A mutual learning framework for few-shot sound event detection,

    D. Yang, H. Wang, Y . Zou, Z. Ye, and W. Wang, “A mutual learning framework for few-shot sound event detection,” inICASSP. IEEE, 2022, pp. 811–815

    2022

  19. [29]

    Active few-shot learning for sound event detection

    Y . Wang, M. Cartwright, and J. P. Bello, “Active few-shot learning for sound event detection.” inINTERSPEECH, 2022, pp. 1551–1555

    2022

  20. [30]

    Few-shot emergency siren detection,

    M. Cantarini, L. Gabrielli, and S. Squartini, “Few-shot emergency siren detection,”Sensors, vol. 22, no. 12, p. 4338, 2022

    2022

  21. [31]

    Learning to detect an animal sound from five examples,

    I. Nolasco, S. Singh, V . Morfi, V . Lostanlen, A. Strandburg-Peshkin, E. Vida ˜na-Vila, L. Gill, H. Pamuła, H. Whitehead, I. Kiskinet al., “Learning to detect an animal sound from five examples,”Ecological informatics, vol. 77, p. 102258, 2023

    2023

  22. [32]

    Real-time automatic drum transcription using dynamic few-shot learning,

    P. Weber, C. Uhle, M. M ¨uller, and M. Lang, “Real-time automatic drum transcription using dynamic few-shot learning,” in2024 IEEE 5th International Symposium on the Internet of Sounds (IS2). IEEE, 2024, pp. 1–8

    2024

  23. [33]

    Interactive singing melody extraction based on active adaptation,

    K. R. Saxena and V . Arora, “Interactive singing melody extraction based on active adaptation,”IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2024

    2024

  24. [34]

    A generalized bayesian model for tracking long metrical cycles in acoustic music signals,

    A. Srinivasamurthy, A. Holzapfel, A. T. Cemgil, and X. Serra, “A generalized bayesian model for tracking long metrical cycles in acoustic music signals,” inICASSP. IEEE, 2016, pp. 76–80

    2016

  25. [35]

    A data-driven bayesian approach to automatic rhythm analysis of indian art music,

    A. Srinivasamurthyet al., “A data-driven bayesian approach to automatic rhythm analysis of indian art music,” 2016

    2016

  26. [36]

    Motif spotting in an alapana in carnatic music

    V . Ishwar, S. Dutta, A. Bellur, and H. A. Murthy, “Motif spotting in an alapana in carnatic music.” inISMIR, 2013, pp. 499–504

    2013

  27. [37]

    Raga verification in carnatic music using longest common segment set

    S. Dutta, K. S. PV , and H. A. Murthy, “Raga verification in carnatic music using longest common segment set.” inISMIR, vol. 1, 2015, pp. 605–611

    2015

  28. [38]

    Sound event detection: A tutorial,

    A. Mesaros, T. Heittola, T. Virtanen, and M. D. Plumbley, “Sound event detection: A tutorial,”IEEE Signal Processing Magazine, vol. 38, no. 5, pp. 67–83, 2021

    2021

  29. [39]

    The performance of sequence alignment algorithms,

    L. Alimehr, “The performance of sequence alignment algorithms,” 2013

    2013

  30. [40]

    Adtof: A large dataset of non-synthetic music for automatic drum transcription,

    M. Zehren, M. Alunno, and P. Bientinesi, “Adtof: A large dataset of non-synthetic music for automatic drum transcription,” inISMIR, 2021, pp. 818–824

    2021

  31. [41]

    Real-time transcription and separation of drum recordings based on nmf decomposition

    C. Dittmar and D. G ¨artner, “Real-time transcription and separation of drum recordings based on nmf decomposition.” inDAFx, 2014, pp. 187– 194

    2014

  32. [42]

    Enst-drums: an extensive audio-visual database for drum signals processing,

    O. Gillet and G. Richard, “Enst-drums: an extensive audio-visual database for drum signals processing,” inISMIR, 2006

    2006

  33. [43]

    Mdb drums: An annotated subset of medleydb for automatic drum transcription,

    C. Southall, C.-W. Wu, A. Lerch, and J. Hockman, “Mdb drums: An annotated subset of medleydb for automatic drum transcription,” 2017

    2017

  34. [44]

    Modal analysis and transcription of strokes of the mridangam using non-negative matrix factorization,

    A. Anantapadmanabhan, A. Bellur, and H. A. Murthy, “Modal analysis and transcription of strokes of the mridangam using non-negative matrix factorization,” inICASSP. IEEE, 2013, pp. 181–185

    2013

  35. [45]

    Tonic- independent stroke transcription of the mridangam,

    A. Anantapadmanabhan, J. Bello, R. Krishnan, and H. Murthy, “Tonic- independent stroke transcription of the mridangam,” inAudio engineer- ing society conference: 53rd international conference: Semantic audio. Audio Engineering Society, 2014

    2014

  36. [46]

    A tutorial on onset detection in music signals,

    J. P. Bello, L. Daudet, S. Abdallah, C. Duxbury, M. Davies, and M. B. Sandler, “A tutorial on onset detection in music signals,”IEEE Transactions on speech and audio processing, vol. 13, no. 5, pp. 1035– 1047, 2005

    2005

  37. [47]

    Drum transcription using partially fixed non-negative matrix factorization,

    C.-W. Wu and A. Lerch, “Drum transcription using partially fixed non-negative matrix factorization,” in23rd European Signal Processing Conference (EUSIPCO). IEEE, 2015, pp. 1281–1285

    2015

  38. [48]

    Hybrid transformers for music source separation,

    S. Rouard, F. Massa, and A. D ´efossez, “Hybrid transformers for music source separation,” inICASSP. IEEE, 2023, pp. 1–5

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.