Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
42
datasets available to search
ShareScore release 0.9.0
Dataset results
42 results for “melody”
Saraga-Carnatic-Melody-Synth
<p>This repository contains the Saraga-Carnatic-Melody-Synth (SCMS), a dataset focusing on time-aligned and continuous vocal melody annotations for the Carnatic music tradition. The annotations have been compiled using an own implemented Analysis/Synthesis framework, and the input data for the framework is the multi-track audio of the <a href="https://zenodo.org/record/1256127">Saraga Dataset</a>.</p><p>This dataset may be used for the research on automatic melody extraction for the Carnatic music tradition.</p><p> </p><p><strong>Dataset contents</strong></p><p>The dataset includes audio excerpts of 30 seconds of the mixtures, together with .csv vocal melody annotations and .lab activations. It also includes metadata files to split between train and test sets, and to relate each excerpt with its respective artists. It also includes relevant metadata about each concert: number of excerpts per concert, gender of the artists, and tonic of the concert. Note that artists and genre are equally distributed on the train and test splits and that data from single artists is never in both sets simultaneously.</p><p> </p><p><strong>Citation</strong></p><p>This dataset is associated with the following paper:</p><p>Plaja-Roglans, G., Nuttall, T., Pearson, L., Serra, X. and Miron, M., 2023. Repertoire-Specific Vocal Pitch Data Generation for Improved Melodic Analysis of Carnatic Music. Transactions of the International Society for Music Information Retrieval, 6(1), p.13–26.DOI: https://doi.org/10.5334/tismir.137</p><p>Check the paper out for further details, benchmarks, and experiments!</p>
Carbon Dioxide Melody
<p>To communicate the urgency of climate change to a wider audience, I converted CO<sub>2</sub> records into music for the piano using the method of sonification. CO<sub>2</sub> measurements recorded by the NOAA Mauna Loa Observatory were linked to musical pitch, driving the sonification. Musical elements were creatively introduced to add nuance and emotion, and they help to emphasize the rising CO<sub>2</sub> melody. The <em>Carbon Dioxide Melody</em> is both easily accessible in audio format and immersively playable as a piano score. This project encourages engagement while conveying in a new way a sense of urgency and gloom towards the causes of climate change. </p>
4 Bars Monophonic Melodies Dataset (Pitch Sequence)
<p>This dataset is designed for applications in music information retrieval, algorithmic composition, and machine learning tasks involving symbolic music data and it consists in a collection of unique 4 bars monophonic melodies represented as MIDI pitch sequences and each accompanied by thirteen attributes obtained with computational methods. The dataset has been generated using the <a href="https://github.com/resolv-libs" target="_blank" rel="noopener">Resolv</a> system's pipelines starting from the full version of the <a href="https://colinraffel.com/projects/lmd/">Lakh MIDI Dataset</a> (a collection of 176,581 unique MIDI files).</p> <p>The dataset has been used to train the models described in the papers:</p> <div> <blockquote> <p>M. Pettenò, A. I. Mezza, and A. Bernardini, "Conditional Diffusion As Latent Constraints for Controllable Symbolic Music Generation", in Proc. of the 26th International Society for Music Information Retrieval Conference (ISMIR 2025), Daejeon, Korea, S ept. 21-25, 2025.</p> <p>M. Pettenò, A. I. Mezza, and A. Bernardini, "On the Joint Minimization of Regularization Loss Functions in Deep Variational Bayesian Methods for Attribute-Controlled Symbolic Music Generation", in Proc. of the 33rd European Signal Processing Conference (EUSIPCO 2025), Palermo, Italy, Sept. 8-12, 2025.</p> </blockquote> </div> <p>The full article of this work also contains all the details on how the attributes have been obtained and on the implementation of the pipelines used for the generation and here it is worth to point out that melodies have been quanized to 4 steps per quarter and only 4/4 time signatures have been considered, hence each melody consists of N = 64 steps where each step is a number in the range [21-108], the MIDI pitches available in a standard piano, or a token in the set {128, 129} for hold note and note off events respectively. No additional performance features (e.g., dynamics, duration, or timing) are included, making this dataset a purely pitch-based collection.</p> <p>Three datasets (train, validation and test) are provided as <a href="https://www.tensorflow.org/tutorials/load_data/tfrecord" target="_blank" rel="noopener">TFRecord</a> file divided into 8 shards that contain the data in the Tensorflow's <a href="https://www.tensorflow.org/api_docs/python/tf/train/SequenceExample">SequenceExample</a> format in which the <em>feature_lists</em> field contains the pitch sequence as a list of integers and the <em>context</em> field its attributes.</p> <p>The table below shows the numbers of unique melodies contained in the three datasets.</p> <table> <tbody> <tr> <td> </td> <td><strong>Train</strong></td> <td><strong>Validation</strong></td> <td><strong>Test</strong></td> </tr> <tr> <td><strong>Total unique melodies</strong></td> <td>10,126,676</td> <td>70,908</td> <td>22,265</td> </tr> </tbody> </table> <p> </p> <p>And here is the list of computed attributes for each melody:</p> <table> <tbody> <tr> <td><strong>Attribute Name</strong></td> <td><strong><em>SequenceExample</em> Context Key</strong></td> <td><strong>Description</strong></td> </tr> <tr> <td>Toussaint Metrical Complexity</td> <td><code>toussaint</code></td> <td>A metric that measures the degree of syncopation in rhythm patterns.</td> </tr> <tr> <td>Note Density</td> <td><code>note_density</code></td> <td>Measures the density of note onsets within the melody.</td> </tr> <tr> <td>Pitch Range</td> <td><code>pitch_range</code></td> <td>An indicator of how wide or narrow the melody is in terms of its pitch content.</td> </tr> <tr> <td>Contour</td> <td><code>contour</code></td> <td>Measures the degree to which the melody moves up or down.</td> </tr> <tr> <td>Note Change Ratio</td> <td><code>note_change_ratio</code></td> <td>The number of note changes normalized to the total number of steps N.</td> </tr> <tr> <td>Dynamic Range</td> <td><code>dynamic_range</code></td> <td>The difference between the maximum and minimum note velocities.</td> </tr> <tr> <td>Longest Repetitive Section</td> <td><code>len_longest_rep_section</code></td> <td>The length of the longest repetitive section in the melody normalized to the total number of steps N. A repetitive section is defined as a note that consecutively repeats at least r = 4 times.</td> </tr> <tr> <td>Repetitive Section Ratio</td> <td><code>repetitive_section_ratio</code></td> <td>The ratio between the total number of repetitive sections and a normalization factor N/r = 64/4 = 16.</td> </tr> <tr> <td>Hold Note Steps Ratio</td> <td><code>ratio_hold_note_steps</code></td> <td>The ratio between the number steps where a note is hold and the total steps N.</td> </tr> <tr> <td>Note Off Steps Ratio</td> <td><code>ratio_note_off_steps</code></td> <td>The ratio between the number steps where no note is played and the total steps N.</td> </tr> <tr> <td>Unique Notes Ratio</td> <td><code>unique_notes_ratio</code></td> <td>The ratio of unique notes is defined with respect to the total number of MIDI pitches considered (88) and the total number of steps N.</td> </tr> <tr> <td>Unique Bigrams Ratio</td> <td><code>unique_bigrams_ratio</code></td> <td>It is the ratio of the unique bigrams in the melody with respect to the total numbers of steps N.</td> </tr> <tr> <td>Unique Trigrams Ratio</td> <td><code>unique_trigrams_ratio</code></td> <td>It is the ratio of the unique trigrams in the melody with respect to the total numbers of steps N.</td> </tr> </tbody> </table> <p> </p> <p>To access the content of a <em>SequenceExample</em> use the <em>tf.io.parse_single_sequence_example</em>, for instance:<br><br></p> <div> <blockquote> <pre>tf.io.parse_single_sequence_example(<br> serialized_example,<br> context_features={<br> "toussaint": tf.io.FixedLenFeature([], dtype=tf.float32, default_value=0),<br> "note_density": tf.io.FixedLenFeature([], dtype=tf.float32, default_value=0),<br> },<br> sequence_features=["pitch_seq"]<br>)</pre> </blockquote> </div>
MDB-melody-synth
<p>MDB-melody-synth<br> ================</p> <p>MDB-melody-synth (c) by Justin Salamon, Rachel Bittner, Jordi Bonada, Juan Jose Bosch, Emilia Gómez and Juan Pablo Bello.<br> MDB-melody-synth is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). <br> You should have received a copy of the license along with this work. If not, see http://creativecommons.org/licenses/by-nc/4.0/</p> <p><br> Created By<br> ----------</p> <p>Justin Salamon*, Rachel Bittner*, Jordi Bonada^, Juan Jose Bosch^, Emilia Gómez^ and Juan Pablo Bello*.<br> * Music and Audio Research Lab (MARL), New York University, USA<br> ^ Music Technology Group, Universitat Pompeu Fabra, Spain<br> http://synthdatasets.weebly.com/<br> http://steinhardt.nyu.edu/marl/<br> https://www.upf.edu/web/mtg</p> <p>Version 1.0.0</p> <p><br> Description<br> -----------</p> <p>MDB-melody-synth contains 65 songs from the MedleyDB dataset (http://medleydb.weebly.com/) in which the melody track <br> has been resynthesized to obtain a perfect melody f0 annotation using the analysis/synthesis method described in the <br> following publication:</p> <p>J. Salamon, R. M. Bittner, J. Bonada, J. J. Bosch, E. Gómez, and J. P. Bello. "An analysis/synthesis framework for <br> automatic f0 annotation of multitrack datasets". In 18th Int. Soc. for Music Info. Retrieval Conf., Suzhou, China, <br> Oct. 2017.</p> <p>This dataset includes:<br> * 65 stereo wav files of song mixes where the original melody stem (track) has been replaced with a resynthesized one<br> * 65 mono wav files containing the resynthesized melody stem (track) only<br> * 65 csv files containing a perfect f0 annotation of the melody stem (track) obtained via the analysis/synthesis <br> method described in the paper</p> <p>The data come in three folders, the contents of which is described below.</p> <p><br> audio_mix<br> ---------<br> Contains 65 stereo wav files of song mixes where the original melody stem (track) has been replaced with a <br> resynthesized one. All other stems (tracks) in the mix are the original ones (not resynthesized). The resynthesized <br> melody is obtained via the analysis/synthesis method described in the paper, and is automatically mixed together <br> with the rest of the stems as described in the paper.</p> <p>Naming convention: <br> <artist>_<songtitle>_MIX_melsynth.wav</p> <p>Example: <br> AClassicEducation_NightOwl_MIX_melsynth.wav</p> <p><br> audio_melody<br> ------------<br> Contains 65 mono wav files of the resynthesized melody stem (track) only. The resynthesized melody is obtained via the <br> analysis/synthesis method described in the paper.</p> <p>Naming convention: <br> <artist>_<songtitle>_STEM_<stemID>.RESYN.wav</p> <p>Example: <br> AClassicEducation_NightOwl_STEM_08.RESYN.wav</p> <p><br> annotation_melody<br> -----------------<br> Contains 65 csv files containing a perfect f0 annotation of the melody stem (track) obtained via the analysis/synthesis <br> method described in the paper. </p> <p>Format:<br> Each file contains two comma-separated columns, the first containing timestamps and the second containing the melody <br> f0 in Hz. The first frame in the annotation is zero-centered. Silence is indicated as 0 Hz. The hop size of the <br> annotation is 128/44100 seconds (~2.9 ms).</p> <p>Naming convention:<br> <artist>_<songtitle>_STEM_<stemID>.RESYN.csv</p> <p>Example:<br> AClassicEducation_NightOwl_STEM_08.RESYN.csv</p> <p><br> Please Acknowledge MDB-melody-synth in Academic Research<br> --------------------------------------------------------</p> <p>Please cite the following publication when using MDB-melody-synth:</p> <p>J. Salamon, R. M. Bittner, J. Bonada, J. J. Bosch, E. Gómez, and J. P. Bello. "An analysis/synthesis framework for <br> automatic f0 annotation of multitrack datasets". In 18th Int. Soc. for Music Info. Retrieval Conf., Suzhou, China, <br> Oct. 2017.</p> <p>For information about the original MedleyDB dataset please see (and cite):</p> <p>R. M. Bittner, J. Salamon, M. Tierney, M. Mauch, C. Cannam, and J. P. Bello. MedleyDB: A multitrack dataset for <br> annotation-intensive MIR research. In 15th Int. Soc. for Music Info. Retrieval Conf., pages 155–160, Taipei, Taiwan, <br> Oct. 2014.</p> <p><br> Conditions of Use<br> -----------------</p> <p>Dataset created by Justin Salamon, Rachel Bittner, Jordi Bonada, Juan Jose Bosch, Emilia Gómez and Juan Pablo Bello. <br> <br> The MDB-melody-synth dataset is offered free of charge under the terms of the Creative Commons<br> Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0): http://creativecommons.org/licenses/by-nc/4.0/<br> <br> The dataset and its contents are made available on an "as is" basis and without warranties of any kind, including <br> without limitation satisfactory quality and conformity, merchantability, fitness for a particular purpose, accuracy or <br> completeness, or absence of errors. Subject to any liability that may not be excluded or limited by law, NYU is not <br> liable for, and expressly excludes, all liability for loss or damage however and whenever caused to anyone by any use of <br> the MDB-melody-synth dataset or any part of it.</p> <p><br> Feedback<br> --------</p> <p>Please help us improve MDB-melody-synth by sending your feedback to: justin.salamon@gmail.com<br> In case of a problem report please include as many details as possible.<br> </p>
MAST melody dataset
<p>The Musical Aptitude Standard Test <strong>(MAST) melody dataset </strong>is designed and shared to facilitate comparison of algorithms in the field of automatic music performance assessment.</p> <p>The dataset includes melodic pattern reproduction performances by students (singing) together with the reference melodic pattern played on piano and assessment results. All recordings are collected during entrance exams (in 2015 and 2016) of Istanbul Technical University (ITU). </p> <p>The recordings are annotations in 2022 within the context of another research project supported by TUBITAK with grant number 121E198 as a part of the Scientific and Technological Research Projects Funding Program (1001). </p> <p>Annotations are performed via blind listening of individual performances after listening to a few renditions of the melodic pattern by the experts. The files were presented in random order (after grouping samples in terms of melodic patterns) (i.e. the expert annotated all samples of a melodic pattern in random order and moved to the next group of samples for the next melodic pattern). </p> <p>A 4-level grading system was used during the evaluations of the data set; <br> 1-Completely Off, 2-Major Mistakes, 3-Minor Mistakes, and 4-Perfect. </p> <p>Annotations were carried by 3 experts; a professor of musicology who has taken part as a jury member in entrance <br> exam auditions, and two music conservatory students of graduate-level programs. The last two annotators re-annotated<br> all collection a few months after the first annotation task. The csv file contains 5 annotations in 5 columns where<br> two of these columns are for the repeated annotations. To facilitate analysis, we also added columns that carry a flag<br> if all annotations match (column: 'fullAgree'), the score/grade all annotations agreed on (column: 'fullAgree_score') <br> and the majority score.</p> <p>In addition to audio files and annotations, two commonly used features are also included: 1) f0-series extracted using Crepe Pitch Tracker (https://github.com/marl/crepe), 2) chroma features computed using Librosa library's chroma_stft function (https://librosa.org/doc/main/generated/librosa.feature.chroma_stft.html)</p> <p>The directory structure is:</p> <ul> <li>annotations: Contains 5 distinct annotations by 3 experts in a scale 1-4 in a csv file. </li> <li>f0data_crepe: MASTmelody dataset, latest version (f0 series extracted using Crepe Pitch Tracker)</li> <li>audioFiles: Audio files sampled at 8kHz</li> <li>chroma: chroma features computed using Librosa library's chroma_stft function using its default settings except n_chroma which is set to 24. </li> </ul> <p>If you use this dataset, please refer to the following paper which announced its original version:</p> <blockquote> <p>Bozkurt, B., Baysal, O., Yuret, D. A Dataset and Baseline System for Singing Voice Assessment, 13th Int. Symposium on Computer Music Multidisciplinary Research, Porto, Sept. 25-28, 2017.</p> </blockquote> <pre><code>@inproceedings{inproceedings, author={Bozkurt, B., Baysal, O., Yuret, D.}, title={A Dataset and Baseline System for Singing Voice Assessment}, year={2017}, booktitle={13th Int. Symposium on Computer Music Multidisciplinary Research, CMMR 2017} }</code></pre>
Melody® Transcatheter Pulmonary Valve Post-Approval Study
ClinicalTrials.gov study NCT01186692. IPD Sharing: UNDECIDED. Countries: 1. Publications: 3.
Melody PB1016 Surveillance Study
ClinicalTrials.gov study NCT02347189. IPD Sharing: Not stated. Countries: 4. Publications: 1.
Melody Transcatheter Pulmonary Valve Study: Post Approval Study of the Investigational Device Exemption Cohort
ClinicalTrials.gov study NCT00740870. IPD Sharing: Not stated. Countries: 1. Publications: 8.
Exploring The Potential of GPT-3-based Large Language Model For Melody Generation
<p>Here, we provide the dataset used, all generated melodies and melodies used to conduct subjective listening test.</p>
ORCHSET: a dataset for melody extraction in symphonic music recordings
<p>Orchset is intended to be used as a dataset for the development and evaluation of melody extraction algorithms. This collection contains 64 audio excerpts focused on symphonic music. with their corresponding annotation of the melody.</p> <p>Melody is here defined as “the single (monophonic) pitch sequence that a listener might reproduce if asked to whistle or hum a piece of polyphonic music”.</p> <p>The dataset creation comprised several tasks: excerpts selection, recording sessions of people singing along with the excerpts, analysis of the recordings and melody annotation. A complete description of the dataset and the creation methodology is presented in this paper:</p> <blockquote> <p>Bosch, J., Marxer, R., Gomez, E., “Evaluation and Combination of Pitch Estimation Methods for Melody Extraction in Symphonic Classical Music”, Journal of New Music Research (2016)</p> </blockquote> <p>Please Acknowledge Orchset in Academic Research</p> <p><strong>Using this dataset</strong></p> <p>When Orchset is used for academic research, we would highly appreciate if scientific publications of works partly based on the Orchset dataset quote the above publication.</p> <p>We are interested in knowing if you find our datasets useful! If you use our dataset please email us at <a href="mailto:mtg-info@upf.edu">mtg-info@upf.edu</a> and tell us about your research.</p> <p> </p> <p><a href="https://www.upf.edu/web/mtg/orchset">https://www.upf.edu/web/mtg/orchset</a></p>
TableS1_Dataset - Rhyme Structure for 25 versions of Sadhukarn main melody collected in Thailand, Cambodia, and Laos: CC0 license
<p>The dataset of rhyme music structure obtained 25 different versions of Sadhukarn main melody collected from Thailand, Laos, Cambodia, and their border areas. Specifically, the material outcomes synthesized by Thai music measure and solfege syllables in Western music notation are applied for setting the rhyme structure format. To define Thai rhyme music structure, a pillar tone or “Look-Tok” in Thai musical terms is a necessary material used for the process. This dataset can be used for a wide spectrum of music analysis.</p> <p> </p>
WuYun: Exploring hierarchical skeleton-guided melody generation using knowledge-enhanced deep learning
<p>Here, we provide the Wikifonia dataset in MIDI format and two sets of MIDI files generated in the experiments. We used MuseScore software (i.e., batch convert) to <strong>synthesize MIDI files into audio files</strong></p>
Melody Transcatheter Pulmonary Valve (TPV) Post-Market Surveillance Study
ClinicalTrials.gov study NCT00688571. IPD Sharing: Not stated. Countries: 6. Publications: 2.
Mindful Melody: Feasibility of Implementing Music Listening on an Inpatient Psychiatric Unit and Its Relation to the Use of As Needed Medications for Acute Agitation
ClinicalTrials.gov study NCT04514432. IPD Sharing: UNDECIDED. Countries: 1. Publications: 1.
Infective Endocarditis in Percutaneous Pulmonary Revalvulation: Comparison Between Melody and Sapien Valves
ClinicalTrials.gov study NCT06186258. IPD Sharing: NO. Countries: 7. Publications: 6.
Figure 9 from: Branch WR, Schmitz A, Lobón-Rovira J, Baptista NL, António T, Conradie W (2021) Rock island melody: A revision of the Afroedura bogerti Loveridge, 1944 group, with descriptions of four new endemic species from Angola. Zoosystematics and Evolution 97(1): 55-82. https://doi.org/10.3897/zse.97.57202
Figure 9 Holotype of Afroedura donveae sp. nov. (PEM R17937) from Omauha Lodge, 15 km south of Tambor, Namibe Province, Angola. Photos: Werner Conradie.
Figure 6 from: Branch WR, Schmitz A, Lobón-Rovira J, Baptista NL, António T, Conradie W (2021) Rock island melody: A revision of the Afroedura bogerti Loveridge, 1944 group, with descriptions of four new endemic species from Angola. Zoosystematics and Evolution 97(1): 55-82. https://doi.org/10.3897/zse.97.57202
Figure 6 Holotype of Afroedura bogerti (AMNH 47841) from Namba (Mombolo), Cuanza-Sul Province, Angola. Photos: Luis M.P. Ceríaco.
Figure 5 from: Branch WR, Schmitz A, Lobón-Rovira J, Baptista NL, António T, Conradie W (2021) Rock island melody: A revision of the Afroedura bogerti Loveridge, 1944 group, with descriptions of four new endemic species from Angola. Zoosystematics and Evolution 97(1): 55-82. https://doi.org/10.3897/zse.97.57202
Figure 5 Live photos of Angolan flat geckos: A.Afroedura bogerti (PEM R24187); B.Afroedura wulfhaackei sp. nov. (paratype PEM R24232); C.Afroedura praedicta sp. nov. (holotype NB 854); D.Afroedura donveae sp. nov. (holotype PEM R17937); E.Afroedura vazpintorum sp. nov. (holotype PEM R24118); F.Afroedura vazpintorum sp. nov. (NB 0745); Photos: A, B, D, E. William R. Branch; C, F. Pedro Vaz Pinto.
Figure 8 from: Branch WR, Schmitz A, Lobón-Rovira J, Baptista NL, António T, Conradie W (2021) Rock island melody: A revision of the Afroedura bogerti Loveridge, 1944 group, with descriptions of four new endemic species from Angola. Zoosystematics and Evolution 97(1): 55-82. https://doi.org/10.3897/zse.97.57202
Figure 8 Holotype of Afroedura praedicta sp. nov. (NB 854) from Serra da Neve, Namibe Province, Angola. Photos: Telmo António.
Figure 4 from: Branch WR, Schmitz A, Lobón-Rovira J, Baptista NL, António T, Conradie W (2021) Rock island melody: A revision of the Afroedura bogerti Loveridge, 1944 group, with descriptions of four new endemic species from Angola. Zoosystematics and Evolution 97(1): 55-82. https://doi.org/10.3897/zse.97.57202
Figure 4 Typical habitat of Angolan Afroedura. A. Namba area (A. bogerti); B. Mt Sandula (Afroedura sp. 5); C. Serra da Neve (A. praedicta sp. nov.); D. Omauha Lodge (A. donveae sp. nov.); E. Farm Mucungo (A. vazpintorum sp. nov.); F. Bimbe (A. vazpintorum sp. nov.). Photos: A. Javier Lobón-Rovira; B, D, E. William R. Branch; C. Pedro Vaz Pinto; F. Ninda L. Baptista.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.