Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Recognizing Ornaments in Vocal Indian Art Music with Active Annotation

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Ornament detection in Indian classical vocals hits 90% F1

desk verdict The ROD dataset is a genuinely useful resource, but the evaluation's test-time chunking consumes ground-truth boundaries, so the reported F1 scores are not a valid measure of detection in continuous audio. read the letter →

arxiv 2505.04419 v1 pith:TGYX6UMF submitted 2025-05-07 eess.AS cs.AIcs.LGcs.SD

Sumit Kumar , Parampreet Singh , Vipul Arora This is my paper · ORCID
classification eess.AScs.AIcs.LGcs.SD
keywords ornamentationdetectionIndianartmusicHindustaniclassicalaudioeventtemporalconvolutionalnetworkactiveannotationchromagramdon'tcarelabelling
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 works to make microtonal vocal ornaments—Kan, Mind, Murki, Nyas, Andolan, and Gamak—machine-recognizable in Hindustani classical vocal music. It introduces the ROD dataset, 4.08 hours of expert-sung studio recordings across four ragas, labelled through a two-stage active-annotation process with a reported inter-annotator agreement of 0.79. The core claim is that ornament detection succeeds when ornaments are treated as events whose full temporal shape is preserved: audio is chunked so that an ornament crossing a boundary is never cut in both chunks, and the cut portion is masked with a "don't care" label during training. On the ROD general split the proposed ED-TCN model reaches 90.06 $F_1$ without a collar and 91.98 with a 200 ms collar, against 74.35 for the CRNN baseline, and after fine-tuning on real-world concert recordings it reaches 67.59 $F_1$ versus 53.68 for a baseline trained from scratch. A sympathetic reader would care because ornamentation is central to Indian art music, not an optional decoration, and this is a working recipe for detecting it automatically.

What carries the argument

The load-bearing mechanism is the interaction between event-aware chunking and masked training. Each 10-second chunk starts at the onset of any ornament that would otherwise be cut by the boundary, and the already-partially-seen tail of a truncated ornament is labelled $\dagger$ ("don't care"); the loss in Eq. (3) then ignores those frames. This is paired with an encoder-decoder temporal convolutional network (ED-TCN), where the encoder applies dilated 1D convolutions with rates 1, 2, 3, 4 and max-pooling, the decoder upsamples symmetrically, and every convolution uses periodic padding that wraps the chroma rows (pitch classes) instead of zero-filling. Periodicity in pitch and dilated receptive fields let the model see a full ornament's temporal structure, while the $\dagger$ mask stops fragmented labels from teaching the model to misclassify partial events.

What would settle it

Take the trained model and score it on the ROD test files cut into fixed 10-second windows with no access to event boundaries, then compare $F_1$ to the event-aware-chunking result; if performance falls toward the baseline, the reported gains are conditional on knowing where ornaments start.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes a working recipe for automatic recognition of vocal ornaments in Hindustani classical music. It contributes the ROD dataset, 212 studio recordings (4.08 hours) by two expert singers across four ragas, strongly labelled for six ornament types with a two-stage active-annotation protocol and a reported Cohen's kappa of 0.79. Its proposed ED-TCN model takes 120-bin chromagrams as input and adds three adaptations to a standard temporal-convolution backbone: "don't care" labels that mask truncated ornament fragments during loss computation, periodic padding that wraps pitch classes instead of zero-padding, and dilated convolutions that enlarge the temporal context. Across general, intra-singer, inter-singer, and raga-specific splits, the model reports $F_1$ between 76.72 and 92.55 without a collar and consistently beats the CRNN baseline; on a separate real-world concert dataset, fine-tuning lifts $F_1$ to 67.59 against 53.68 for a from-scratch baseline. The ablation study identifies "don't care" labelling as the largest single contributor, worth 7–8 points of $F_1$ over the same model without it.

Load-bearing premise

The reported results assume the test audio is chunked with knowledge of the true ornament boundaries, so a truncated ornament never appears at test time; the paper does not describe how the model would be applied to long recordings when those boundaries are unknown.

Editorial extensions

If this is right

  • At 90.06 $F_1$ on the general ROD split (91.98 with a 200 ms collar), the model is a usable front end for music-pedagogy feedback, singer-identification features, or expressive-synthesis conditioning, tasks the paper lists as downstream motivations.
  • Inter-singer experiments keep $F_1$ above 80 without a collar when training on one singer and testing on the other, implying the detector learns ornament structure rather than singer-specific timbre.
  • Raga-specific experiments stay near or above 75 $F_1$, and the concert-recording experiment reaches 67.59 after fine-tuning, so the approach extends beyond studio lessons to broadcast-quality archival audio after vocal-stem separation.
  • Because "don't care" labelling contributes more than any other component, the paper's recipe implies that future ornament datasets should be chunked with event boundaries in mind from the start, not with fixed windows.

Reading between the lines

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

  • Inference beyond the paper: the "don't care" masking idea is transportable to any event-detection task whose events are identified by their internal temporal shape—animal vocalizations, drum strokes, coughs—where fixed-window cropping corrupts the evidence.
  • Inference beyond the paper: a field-ready system would need a label-free way to choose chunk starts on long audio, since the paper's chunking procedure uses ground-truth onsets; a natural extension is a boundary-proposal network that predicts event starts and then feeds those chunks to the detector.
  • Inference beyond the paper: the same two-stage active-annotation workflow could produce comparable ornament datasets for other oral music traditions, with the event label scheme adapted to each tradition's ornament inventory.
  • Inference beyond the paper: the confusion patterns (Kan–Mind, Murki–Kan, Andolan–Mind) suggest that adding raga or gharana metadata as conditioning input could break the structurally ambiguous cases the current model still merges.
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 / 5 minor

Summary. The paper introduces ROD, a new dataset of expert-annotated vocal ornaments in Hindustani classical music, and proposes an ED-TCN model with don't-care labeling, periodic padding, and dilated convolutions for frame-wise ornament detection. The authors report F1 improvements over a CRNN baseline across several train-test configurations (Table III) and on a separate Prasar Bharati concert dataset (Table IV). The central claim is that the proposed model outperforms the baseline and generalizes across singers, ragas, and real-world concert audio.

Significance. The ROD dataset is a potentially valuable community resource: it contains 4.08 hours of expert-annotated recordings with six ornament classes, reports inter-annotator agreement (Cohen's Kappa 0.79), and the authors state that code, models, and demos will be released. The ablation study isolating the contributions of don't-care labeling, periodic padding, dilated convolutions, and chroma resolution is informative. However, the evaluation protocol has a load-bearing flaw: test-time chunking uses ground-truth event boundaries, so the reported generalization scores are not currently supported. If the authors provide a label-free inference protocol and re-run the experiments with proper splits and a fairly trained baseline, the paper could make a substantial contribution.

major comments (4)
  1. [V-A and VI] The chunking procedure defined in Section V-A uses ground-truth event onsets and offsets (o_ij, f_ij) to decide chunk starts and to assign 'don't care' labels to truncated tails. Section VI states that all audio is segmented into 10-second chunks using this technique, which necessarily includes the test splits used in Tables III and IV. At inference time on unlabeled audio, event onsets and offsets are unknown, and the paper does not describe any label-free inference protocol (fixed windows, sliding windows, or a separate event detector). Consequently, the reported F1 scores are produced under an oracle segmentation that consumes target labels. This invalidates the claims of generalization in Sections VII-A and VII-B. The authors must specify how test audio is chunked during inference and re-evaluate with a protocol that does not use ground-truth boundaries. In addition, the treatment of 'don't care' frames during test evaluation is unspecified; if truncated tails are marked as 'don't care' in test chunks, the paper must state whether those frames are excluded from the precision/recall computation.
  2. [VII-A1 and Table II] Experiment 1 is described as a 'random 70/20/10 train-test-validation split,' but the paper does not state whether the split is at the file level or the chunk level. The chunking technique in Section V-A produces overlapping chunks when an event crosses a boundary, so a chunk-level split can place overlapping chunks from the same audio file into both training and test sets, causing information leakage and inflating the reported F1. This ambiguity must be resolved; if the split was chunk-level, the experiments should be rerun with file-level splits.
  3. [VI-A and VI-B] The baseline CRNN is trained for 100 epochs, while the proposed ED-TCN is trained for 3000 epochs. This 30-fold difference in training budget means the reported superiority in Table III could be attributed to under-training the baseline rather than to the architectural benefits of the proposed model. The authors should train the baseline to convergence (e.g., with early stopping or a comparable number of epochs) and report learning curves or a convergence criterion for both models.
  4. [VII-B and Table IV] The comparison on the Prasar Bharati dataset is not apples-to-apples: the 'Proposed Model (with FT)' is pre-trained on ROD and then fine-tuned, whereas the baseline is trained from scratch on the Prasar Bharati data. Any advantage may reflect the benefit of pre-training rather than the model design. The authors should compare against a baseline that is also pre-trained on ROD and fine-tuned, or report the proposed model trained from scratch on Prasar Bharati, to isolate the architectural contribution.
minor comments (5)
  1. [V-A, equations] In the periodic padding description, 'Xbottom = X[:,p]' appears to be a typo; it should likely be 'Xbottom = X[:p]' (first p rows). Please correct the notation.
  2. [VI-C] The paper states that 'Precision, Recall, and F1-score averaged over the classes' are reported, but it does not specify whether the averaging is macro (unweighted per class) or micro (weighted by support). This should be stated for reproducibility.
  3. [Table II caption] The abbreviations 'Str1' and 'Ste1' are not defined in the caption or text; please define them as 'Training subset of Singer 1' and 'Testing subset of Singer 1' explicitly.
  4. [VII-C] The text claims that don't care labelling 'consistently boosts F1 scores by 7–8%', but Table V shows an increase from 79.02 (without†) to 90.06 (with†), which is about 11 points. The text also says 'provides an increased performance of 11%' a few sentences earlier. Please reconcile these numbers.
  5. [Abstract and I] The GitHub URL in the abstract and introduction is given as 'https://github.com/madhavlab/2024 ornamentation' with a space; this will render as an invalid link. Please provide the correct URL.

Circularity Check

1 steps flagged · score 4.0 of 10

Reported generalization F1 is measured on test chunks constructed from ground-truth event boundaries; no label-free inference procedure is specified, so the absolute deployment claim is not yet supported, though the relative model comparison and ablations remain internally consistent.

  1. other [Section V-A (Data Preprocessing) and Section VI (Experiments)]
    "if oij≤tik+t<fij, then ti(k+1)=oij. ... The set (X,Y) is used for training and testing the model. ... Since the audio files have variable lengths, we segment them into 10-second chunks using the chunking technique discussed in Section V-A."

    The chunking rule moves the next chunk start to the labeled onset of any event that crosses a boundary, so test inputs are functions of ground-truth labels. Section VI applies this same label-dependent chunking to all audio, including the test splits behind the headline F1 scores (90.06/91.98 on ROD, 67.59 on Prasar Bharati). On unlabeled concert audio there is no described way to choose chunk starts from event onsets, so the reported numbers are achieved under an oracle segmentation that also removes boundary-truncated frames from consideration. This makes the absolute generalization claim circular in an evaluation-protocol sense: the test setting presupposes the very event boundaries the system is supposed to detect.

full rationale

The model itself is not derived from its labels: ED-TCN is an independently specified architecture, and the ablations (don't-care labelling, periodic padding, dilated convolutions) are defined and compared under the same protocol. The CRNN baseline is also chunked the same way, so the relative ranking is internally consistent. The main circularity-like defect is that the chunking technique in V-A consumes ground-truth onsets/offsets and Section VI applies it to test audio, with no label-free inference procedure stated. That is an evaluation-procedure reduction rather than an equation-level equivalence or a fitted parameter renamed as a prediction; accordingly it does not warrant a high score. Self-citations ([9], [15], [27]) are peripheral and not load-bearing. A remaining ambiguity is whether the Experiment 1 random split is file-level or chunk-level, which could add leakage, but the paper does not specify enough to call it circular.

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

The central contribution rests on the annotation protocol and model design choices. The annotation duration thresholds define the ground truth, and the model hyperparameters are selected by hand. The evaluation additionally assumes the event-aware chunking can be applied at test time without labels, which is not stated.

free parameters (5)
  • Chunk duration t = 10 seconds
    Chosen for all experiments; controls how often ornaments cross chunk boundaries and how much don't-care masking is used.
  • Chroma resolution F = 120 bins (12-bin variant tested)
    The authors select 120 bins after ablation because it gives higher F1; this is a model-selection choice.
  • Evaluation collar = 200 ms
    Adopted from prior work; directly inflates F1 by forgiving boundary errors in event detection.
  • Annotation duration thresholds = Kan <=0.35s; Mind >=0.45s; Nyas >=0.6s; Murki 0.4-1.0s; Andolan >=1.0s; Gamak >=0.7s
    These hand-set rules define the ground-truth classes partly by duration, so the model is trained to reproduce arbitrary thresholds.
  • Model hyperparameters = L=4, filters [32,64,128,256], dilation [1,2,3,4], kernel 5, dropout 0.3, lr 0.001, 3000 epochs
    Hand-selected architecture and training choices; no hyperparameter search or sensitivity analysis is reported.
assumptions (5)
  • domain assumption Ornaments are discrete, non-overlapping events with objectively identifiable onsets and offsets.
    Section IV-A labels all ornaments as strong, non-overlapping events; if boundaries are inherently ambiguous, the ground truth and evaluation are not well-defined.
  • domain assumption Six ornament classes are discriminable from 10-second chromagram context.
    The pipeline uses chromagrams and 10-second chunks throughout; no alternative features are tested.
  • domain assumption Two expert singers in four ragas are sufficient to benchmark ornament detection for Indian Art Music.
    The dataset has only two singers and four ragas; claims of generalization to IAM rest on this small sample.
  • standard math Background audio processing, including STFT, chromagram extraction, and DEMUCS source separation, is correct and standard.
    Used without proof; these are standard tools in the field.
  • domain assumption The Prasar Bharati dataset annotations are reliable.
    No annotation protocol is given for this external set; the generalization claim depends on its correctness.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Recognizing Ornaments in Vocal Indian Art Music with Active Annotation." pith.science (2026). https://pith.science/paper/TGYX6UMF

@misc{pith2026250504419,
  author       = {Pith},
  title        = {Pith review of: Recognizing Ornaments in Vocal Indian Art Music with Active Annotation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TGYX6UMF}},
  note         = {Machine review of arXiv:2505.04419}
}
read the original abstract

Ornamentations, embellishments, or microtonal inflections are essential to melodic expression across many musical traditions, adding depth, nuance, and emotional impact to performances. Recognizing ornamentations in singing voices is key to MIR, with potential applications in music pedagogy, singer identification, genre classification, and controlled singing voice generation. However, the lack of annotated datasets and specialized modeling approaches remains a major obstacle for progress in this research area. In this work, we introduce R\=aga Ornamentation Detection (ROD), a novel dataset comprising Indian classical music recordings curated by expert musicians. The dataset is annotated using a custom Human-in-the-Loop tool for six vocal ornaments marked as event-based labels. Using this dataset, we develop an ornamentation detection model based on deep time-series analysis, preserving ornament boundaries during the chunking of long audio recordings. We conduct experiments using different train-test configurations within the ROD dataset and also evaluate our approach on a separate, manually annotated dataset of Indian classical concert recordings. Our experimental results support the superior performance of our proposed approach over the baseline CRNN.

Figures

Figures reproduced from arXiv: 2505.04419 by the authors.

Figure 1
Figure 1. Chromagram (top) and Pitch Contour (bottom) representations of ornaments in IAM (a): [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Class-wise ornament frequency for Singer 1 (left) and Singer 2 (right), with ornament classes on the x-axis and log-scaled counts on the y-axis. K, [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. An illustration of audio chunking using overlapping windows and [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: An overview of the proposed methodology for the ornamentation detection task [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Encoder-Decoder Temporal Convolutional Network [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Confusion matrix illustrating the performance of the proposed model [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Comparison of (a) 12-dimensional and (b) 120-dimensional chroma [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: The figure depicts an instance of Andolan with a cycle (marked in red lines) resembling M¯ind. The x-axis represents time (in sec). D. Subjective Analysis This section outlines the key sources of error in the predic￾tions of our model and highlights the underlying comp…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Pitch Contour Tokenization using VQ-VAE and Its Application on Korean Traditional Music Analysis

    cs.SD 2026-08 conditional novelty 6.0 of 10

    A VQ-VAE with alignment-invariant reconstruction learns stable tokens of pitch-contour shapes from unlabeled audio and shows modest signal for sigimsae categories and pansori modes.

Reference graph

Works this paper leans on

44 extracted references · 35 canonical work pages · cited by 1 Pith paper

  1. [1]

    A machine learning approach to ornamentation modeling and synthesis in jazz guitar,

    S. Giraldo and R. Ram ´ırez-Melendez, “A machine learning approach to ornamentation modeling and synthesis in jazz guitar,” in Machine Learning and Music Generation . Routledge, 2018, pp. 21–40

    work page 2018

  2. [2]

    Ornamenta- tion in hindustani vocal music,

    K. Banerjee, A. Patranabis, R. Sengupta, and D. Ghosh, “Ornamenta- tion in hindustani vocal music,” in Proceedings of 27th International Symposium on Frontiers of Research in Speech and Music , K. Hirose, D. Joshi, and S. Sanyal, Eds. Singapore: Springer Nature Singapore, 2024, pp. 125–135

    work page 2024

  3. [3]

    Analysis and detection of singing techniques in repertoires of j-pop solo singers,

    Y . Yamamoto, J. Nam, and H. Terasawa, “Analysis and detection of singing techniques in repertoires of j-pop solo singers,” in ISMIR, 2022

    work page 2022

  4. [4]

    Primadnn: A characteristics-aware dnn customization for singing technique detection,

    ——, “Primadnn: A characteristics-aware dnn customization for singing technique detection,” in EUSIPCO. IEEE, 2023, pp. 406–410

    work page 2023

  5. [5]

    Analysis and classification of ornaments in north indian (hindustani) classical music,

    H. Purwins, “Analysis and classification of ornaments in north indian (hindustani) classical music,” Ph.D. dissertation, Master’s thesis, Uni- versitat Pompeu Fabra, Barcelona, Spain, 2010

    work page 2010

  6. [6]

    Deformable cnn and imbalance-aware feature learning for singing technique classification,

    Y . Yamamoto, J. Nam, and H. Terasawa, “Deformable cnn and imbalance-aware feature learning for singing technique classification,” arXiv preprint arXiv:2206.12230 , 2022

    arXiv 2022

  7. [7]

    Convolutional recurrent neural networks for polyphonic sound event detection,

    E. Cakır, G. Parascandolo, T. Heittola, H. Huttunen, and T. Virtanen, “Convolutional recurrent neural networks for polyphonic sound event detection,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 25, no. 6, pp. 1291–1303, 2017

    work page 2017

  8. [8]

    R-crnn: Region-based convolutional recurrent neural network for audio event detection,

    C.-C. Kao, W. Wang, M. Sun, and C. Wang, “R-crnn: Region-based convolutional recurrent neural network for audio event detection,” arXiv preprint arXiv:1808.06627, 2018

    arXiv 2018

Show all 44 references
  1. [9]

    Automatic detection and analysis of singing mistakes for music pedagogy,

    V . Arora, S. Jaiswal, A. Raina, and S. Kumar, “Automatic detection and analysis of singing mistakes for music pedagogy,” Authorea Preprints, 2023

    2023

  2. [10]

    Artificial intelligence-assisted music educa- tion: A critical synthesis of challenges and opportunities,

    J. F. Merch ´an S ´anchez-Jara, S. Gonz ´alez Guti ´errez, J. Cruz Rodr ´ıguez, and B. Syroyid Syroyid, “Artificial intelligence-assisted music educa- tion: A critical synthesis of challenges and opportunities,” Education Sciences, vol. 14, no. 11, p. 1171, 2024

    2024

  3. [11]

    Artificial intelligence in music education: A critical review,

    S. Holland, “Artificial intelligence in music education: A critical review,” Readings in music and artificial intelligence , pp. 239–274, 2013

    2013

  4. [12]

    Sintechsvs: A singing technique controllable singing voice synthesis system,

    J. Zhao, L. Q. H. Chetwin, and Y . Wang, “Sintechsvs: A singing technique controllable singing voice synthesis system,” IEEE/ACM Transactions on Audio, Speech, and Language Processing , vol. 32, pp. 2641–2653, 2024

    2024

  5. [13]

    Singing voice conversion with disentangled representations of singer and vocal tech- nique using variational autoencoders,

    Y .-J. Luo, C.-C. Hsu, K. Agres, and D. Herremans, “Singing voice conversion with disentangled representations of singer and vocal tech- nique using variational autoencoders,” in ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)...

    2020

  6. [14]

    Robust singer identification of indian playback singers,

    D. Y . Loni and S. Subbaraman, “Robust singer identification of indian playback singers,” EURASIP Journal on Audio, Speech, and Music Processing, vol. 2019, pp. 1–14, 2019

    2019

  7. [15]

    Explainable deep learning analysis for raga identification in indian art music,

    P. Singh and V . Arora, “Explainable deep learning analysis for raga identification in indian art music,” arXiv preprint arXiv:2406.02443 , 2024

    2024 arXiv

  8. [16]

    Breathy, resonant, pressed–automatic detection of phonation mode from audio recordings of singing,

    P. Proutskova, C. Rhodes, T. Crawford, and G. Wiggins, “Breathy, resonant, pressed–automatic detection of phonation mode from audio recordings of singing,” Journal of New Music Research , vol. 42, no. 2, pp. 171–186, 2013

    2013

  9. [17]

    Phonation mode detection in singing: A singer adapted model,

    Y . Wang, W. Wei, and Y . Wang, “Phonation mode detection in singing: A singer adapted model,” in ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2023, pp. 1–5

    2023

  10. [18]

    V ocalset: A singing voice dataset

    J. Wilkins, P. Seetharaman, A. Wahl, and B. Pardo, “V ocalset: A singing voice dataset.” in ISMIR, 2018, pp. 468–474

    2018

  11. [19]

    Semantic tagging of singing voices in popular music recordings,

    K. L. Kim, J. Lee, S. Kum, C. L. Park, and J. Nam, “Semantic tagging of singing voices in popular music recordings,” IEEE/ACM Transactions on Audio, Speech, and Language Processing , vol. 28, pp. 1656–1668, 2020

    2020

  12. [20]

    Paralinguistic singing attribute recognition using supervised machine learning for describing the classical tenor solo singing voice in vocal pedagogy,

    Y . Xu, W. Wang, H. Cui, M. Xu, and M. Li, “Paralinguistic singing attribute recognition using supervised machine learning for describing the classical tenor solo singing voice in vocal pedagogy,” EURASIP Journal on Audio, Speech, and Music Processing , vol. 2022, no. 1, p. 8, 2022

    2022

  13. [21]

    Scream detection in heavy metal music,

    V . Kalbag and A. Lerch, “Scream detection in heavy metal music,” arXiv preprint arXiv:2205.05580, 2022

    2022 arXiv

  14. [22]

    Dcase 2017 challenge setup: Tasks, datasets and baseline system,

    A. Mesaros, T. Heittola, A. Diment, B. Elizalde, A. Shah, E. Vincent, B. Raj, and T. Virtanen, “Dcase 2017 challenge setup: Tasks, datasets and baseline system,” in DCASE 2017-workshop on detection and classification of acoustic scenes and events , 2017

    2017

  15. [23]

    A survey of vision-based meth- ods for action representation, segmentation and recognition,

    D. Weinland, R. Ronfard, and E. Boyer, “A survey of vision-based meth- ods for action representation, segmentation and recognition,” Computer vision and image understanding , vol. 115, no. 2, pp. 224–241, 2011

    2011

  16. [24]

    Environmental sound classification with convolutional neural networks,

    K. J. Piczak, “Environmental sound classification with convolutional neural networks,” in 2015 IEEE 25th international workshop on machine learning for signal processing (MLSP) . IEEE, 2015, pp. 1–6

    2015

  17. [25]

    Esc: Dataset for environmental sound classification,

    ——, “Esc: Dataset for environmental sound classification,” in Proceed- ings of the 23rd ACM international conference on Multimedia, 2015, pp. 1015–1018

    2015

  18. [26]

    Deep convolutional neural networks and data augmentation for environmental sound classification,

    J. Salamon and J. P. Bello, “Deep convolutional neural networks and data augmentation for environmental sound classification,” IEEE Signal processing letters, vol. 24, no. 3, pp. 279–283, 2017

    2017

  19. [27]

    Balanced deep cca for bird vocalization detection,

    S. Kumar, B. Anshuman, L. R ¨uttimann, R. H. Hahnloser, and V . Arora, “Balanced deep cca for bird vocalization detection,” in ICASSP 2023- 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2023, pp. 1–5

    2023

  20. [28]

    Segmentation and recognition of tabla strokes

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

    2005

  21. [29]

    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

    2018

  22. [30]

    Temporal action segmentation: An analysis of modern techniques,

    G. Ding, F. Sener, and A. Yao, “Temporal action segmentation: An analysis of modern techniques,” IEEE Transactions on Pattern Analysis and Machine Intelligence , 2023

    2023

  23. [31]

    Temporal convolutional networks: A unified approach to action segmentation,

    C. Lea, R. Vidal, A. Reiter, and G. D. Hager, “Temporal convolutional networks: A unified approach to action segmentation,” in Computer Vision–ECCV 2016 Workshops: Amsterdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part III 14 . Springer, 2016, pp. 47–54. J...

    2016

  24. [32]

    Tempo- ral convolutional networks for action segmentation and detection,

    C. Lea, M. D. Flynn, R. Vidal, A. Reiter, and G. D. Hager, “Tempo- ral convolutional networks for action segmentation and detection,” in proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 156–165

    2017

  25. [33]

    Audacity: Free audio editor and recorder,

    The Audacity Team, “Audacity: Free audio editor and recorder,” 2024. [Online]. Available: https://www.audacityteam.org/

    2024

  26. [34]

    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

  27. [35]

    The gharana system: Lineage in indian classical music,

    J. Harrison, “The gharana system: Lineage in indian classical music,” 2023, accessed: 2025-02-22. [Online]. Available: https://www.darbar. org/article/the-gharana-system-lineage-in-indian-classical-music

    2023

  28. [36]

    Fsd50k: An open dataset of human-labeled sound events,

    E. Fonseca, X. Favory, J. Pons, F. Font, and X. Serra, “Fsd50k: An open dataset of human-labeled sound events,” October 2020. [Online]. Available: https://doi.org/10.5281/zenodo.4060432

    2020 doi

  29. [37]

    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,” in2015 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, 2015, pp. 5206–5210

    2015

  30. [38]

    A classification approach to melody transcription

    G. E. Poliner and D. P. Ellis, “A classification approach to melody transcription.” in ISMIR, vol. 2005, 2005, p. 6th

    2005

  31. [39]

    M ¨uller, Fundamentals of music processing: Using Python and Jupyter notebooks

    M. M ¨uller, Fundamentals of music processing: Using Python and Jupyter notebooks. Springer, 2021, vol. 2

    2021

  32. [40]

    Ms-tcn: Multi-stage temporal convolutional network for action segmentation,

    Y . A. Farha and J. Gall, “Ms-tcn: Multi-stage temporal convolutional network for action segmentation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 3575– 3584

    2019

  33. [41]

    Towards understanding action recognition,

    H. Jhuang, J. Gall, S. Zuffi, C. Schmid, and M. J. Black, “Towards understanding action recognition,” in Proceedings of the IEEE interna- tional conference on computer vision , 2013, pp. 3192–3199

    2013

  34. [42]

    Circular convo- lutional neural networks for panoramic images and laser data,

    S. Schubert, P. Neubert, J. P ¨oschmann, and P. Protzel, “Circular convo- lutional neural networks for panoramic images and laser data,” in 2019 IEEE intelligent vehicles symposium (IV) . IEEE, 2019, pp. 653–660

    2019

  35. [43]

    Wavenet: A generative model for raw audio,

    A. van den Oord, “Wavenet: A generative model for raw audio,” arXiv preprint arXiv:1609.03499, 2016

    2016 arXiv

  36. [44]

    Hybrid spectrogram and waveform source separation,

    A. D ´efossez, “Hybrid spectrogram and waveform source separation,” arXiv preprint arXiv:2111.03600 , 2021

    2021 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.