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.
112
datasets available to search
ShareScore release 0.9.0
Dataset results
112 results for “music dataset”
Dataset: Warner Music Group Corp. (WMG) Stock Performance
This dataset provides historical stock market performance data for specific companies. It enables users to analyze and understand the past trends and fluctuations in stock prices over time. This information can be utilized for various purposes such as investment analysis, financial research, and market trend forecasting.
Turkish Makam Music Audio-Score Alignment Dataset
<p>This release contains the annotations and the scores to test the audio-score alignment methodology explained in:</p> <blockquote> <p><em>Şentürk, S., Gulati, S., and Serra, X. (2014). <strong>Towards alignment of score and audio recordings of Ottoman-Turkish makam music.</strong> In Proceedings of 4th International Workshop on Folk Music Analysis, pages 57–60, Istanbul, Turkey.</em></p> </blockquote> <p>The dataset in this release is derived from the transcription test dataset used in the paper:</p> <blockquote> <p><em>Benetos, E. & Holzapfel, A. (2013). <strong>Automatic transcription of Turkish makam music.</strong> In Proceedings of 14th International Society for Music Information Retrieval Conference, 4 - 8 Nov 2013, Curitiba, PR, Brazil.</em></p> </blockquote> <p>The scores for each composition are obtained from the SymbTr collection explained in:</p> <blockquote> <p><em>Karaosmanoğlu, K. (2012). <strong>A Turkish makam music symbolic database for music information retrieval: SymbTr.</strong> In Proceedings of 13th International Society for Music Information Retrieval Conference (ISMIR), pages 223–228.</em></p> </blockquote> <p>From the annotated score onsets for some of the above recordings only the main singing voice segments have been selected. Further separately only a subset of vocal onsets crresponding to phoneme transitions rules have been explicitly annotated as annotationOnsets.txt</p> <blockquote> <p><a href="http://mtg.upf.edu/biblio/author/810">Dzhambazov, G.</a>, <a href="http://mtg.upf.edu/biblio/author/644">Srinivasamurthy A.</a>, <a href="http://mtg.upf.edu/biblio/author/494">Şentürk S.</a>, & <a href="http://mtg.upf.edu/biblio/author/1012">Serra X.</a> (2016). <a href="http://mtg.upf.edu/node/3492">On the Use of Note Onsets for Improved Lyrics-to-audio Alignment in Turkish Makam Music</a>. 17th International Society for Music Information Retrieval Conference (ISMIR 2016</p> </blockquote> <p><strong>Using this dataset</strong></p> <p>Please cite the above publications if you use this dataset in a 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="http://compmusic.upf.edu/node/233">http://compmusic.upf.edu/node/233 </a></p>
Dataset and software for "Optimizing beat synchronized running to music"
<p>These are the dataset and Matlab scripts that were used for analysis as described in the publication on "<strong>Optimizing beat synchronized running to music"</strong></p>
#nowplaying-RS: A New Benchmark Dataset for Building Context-Aware Music Recommender Systems
<p>Music recommender systems can offer users personalized and contextualized recommendation and are therefore important for music information retrieval. An increasing number of datasets have been compiled to facilitate research on different topics, such as content-based, context-based or next-song recommendation. However, these topics are usually addressed separately using different datasets, due to the lack of a unified dataset that contains a large variety of feature types such as item features, user contexts, and timestamps. To address this issue, we propose a large-scale benchmark dataset called #nowplaying-RS, which contains 11.6 million music listening events (LEs) of 139K users and 346K tracks collected from Twitter. The dataset comes with a rich set of item content features and user context features, and the timestamps of the LEs. Moreover, some of the user context features imply the cultural origin of the users, and some others—like hashtags—give clues to the emotional state of a user underlying an LE. In this paper, we provide some statistics to give insight into the dataset, and some directions in which the dataset can be used for making music recommendation. We also provide standardized training and test sets for experimentation, and some baseline results obtained by using factorization machines.</p> <p>The dataset contains three files:</p> <ul> <li>user_track_hashtag_timestamp.csv contains basic information about each listening event. For each listening event, we provide an id, the user_id, track_id, hashtag, created_at </li> <li>context_content_features.csv: contains all context and content features. For each listening event, we provide the id of the event, user_id, track_id, artist_id, content features regarding the track mentioned in the event (instrumentalness, liveness, speechiness, danceability, valence, loudness, tempo, acousticness, energy, mode, key) and context features regarding the listening event (coordinates (as geoJSON), place (as geoJSON), geo (as geoJSON), tweet_language, created_at, user_lang, time_zone, entities contained in the tweet).</li> <li>sentiment_values.csv contains sentiment information for hashtags. It contains the hashtag itself and the sentiment values gathered via four different sentiment dictionaries: AFINN, Opinion Lexicon, Sentistrength Lexicon and vader. For each of these dictionaries we list the minimum, maximum, sum and average of all sentiments of the tokens of the hashtag (if available, else we list empty values). However, as most hashtags only consist of a single token, these values are equal in most cases. Please note that the lexica are rather diverse and therefore, are able to resolve very different terms against a score. Hence, the resulting csv is rather sparse. The file contains the following comma-separated values: <hashtag, vader_min, vader_max, vader_sum,vader_avg, afinn_min, afinn_max, afinn_sum, afinn_avg, ol_min, ol_max, ol_sum, ol_avg, ss_min, ss_max, ss_sum, ss_avg >, where we abbreviate all scores gathered over the Opinion Lexicon with the prefix 'ol'. Similarly, 'ss' stands for SentiStrength. </li> </ul> <p>Please also find the training and test-splits for the dataset in this repo. Also, prototypical implementations of a context-aware recommender system based on the dataset can be found at <a href="https://github.com/asmitapoddar/nowplaying-RS-Music-Reco-FM">https://github.com/asmitapoddar/nowplaying-RS-Music-Reco-FM</a>.</p> <p>If you make use of this dataset, please cite the following paper where we describe and experiment with the dataset:</p> <p>@inproceedings{smc18,<br> title = {#nowplaying-RS: A New Benchmark Dataset for Building Context-Aware Music Recommender Systems},<br> author = {Asmita Poddar and Eva Zangerle and Yi-Hsuan Yang},<br> url = {http://mac.citi.sinica.edu.tw/~yang/pub/poddar18smc.pdf},<br> year = {2018},<br> date = {2018-07-04},<br> booktitle = {Proceedings of the 15th Sound & Music Computing Conference},<br> address = {Limassol, Cyprus},<br> note = {code at https://github.com/asmitapoddar/nowplaying-RS-Music-Reco-FM},<br> tppubtype = {inproceedings}<br> }</p>
Arab-Andalusian music lyrics dataset
<p>The dataset contains lyrics for the songs in the <a href="https://musicbrainz.org/collection/142ea0d7-7fdf-4ea5-9b04-219f68023d01">Arab-Anadalusian music collection curated within the CompMusic project</a>, that belong to the nawbas "Isbahan", "Maya”, “Raml Maya”, “Gharibat al-Husayn”, “Hijaz Kabir”, “Hijaz Msharqi”, “Istihlal”, “Rasd”, and ”Rasd Dayl”.</p> <p>Lyrics are stored in two formats: as Tab Separated Values (TSV) files and as JSON files.</p> <p>Each file is identified by its MusicBrainz recording ID (MBID).</p> <p>The lyrics are stored both in their original Arabic script (folder 'original') and a romanized/transliterated version (folder 'transliterated') using the American Library of Congress (ALA-LC standard).</p> <p>Corresponding audio files are available from <a href="https://zenodo.org/record/1291776#.WyeYnZ9fjCI">the Arab-Andalusian music corpus</a>, as well as the Internet Archive URL included in the metadata file ('metadata.csv').</p> <p>For more information about the exact format and contents of the dataset, please consult the README provided in the archive.</p> <p>For more information, please refer to <a href="http://compmusic.upf.edu/corpora">http://compmusic.upf.edu/corpora</a>.</p>
SynthSOD: Developing an Heterogeneous Dataset for Orchestra Music Source Separation
<p>The SynthSOD dataset contains more than 47 hours of multitrack music obtained by synthesizing orchestra and ensemble pieces from the <a href="https://qsdfo.github.io/LOP/database.html" target="_blank" rel="noopener">Symbolic Orchestral Database (SOD)</a> using Spitfire BBC Symphony Orchestra Professional Library. To synthesize the MIDI files from the SOD, we needed to fix the original files into the General MIDI standard, select a subsect of files that fitted into our requirements (e.g., containing only instruments that we could synthesize), and develop a new system to generate musically-motivated random annotations about tempo, dynamic, and articulation. The code to replicate this process is available in <a href="https://github.com/repertorium/HQ-SOD-generator" target="_blank" rel="noopener">our repository</a> and all the details can be read in <a href="https://doi.org/10.1109/OJSP.2025.3528361" target="_blank" rel="noopener">our paper</a>. We have also published the code to train and evaluate the baseline and the pre-trained models in a <a href="https://github.com/repertorium/SynthSOD-Baseline" target="_blank" rel="noopener">GitHub repository</a>.</p> <p>We have also published the aligned score information for most of the pieces <a href="https://doi.org/10.5281/zenodo.14971533">here</a>.</p>
Datasets from the KDD 2021 article "A Semi-Personalized System for User Cold Start Recommendation on Music Streaming Apps"
<p>We publicly release the anonymized <em>song_embeddings.parquet user_embeddings.parquet user_features_test.parquet user_features_train.parquet user_features_validation.parquet</em> datasets, with each of the TT-SVD or UT-ALS versions of embeddings, from the music streaming platform Deezer, as described in the article "<em>A Semi-Personalized System for User Cold Start Recommendation on Music Streaming Apps"</em> published in the proceedings of the 27TH ACM SIGKDD conference on knowledge discovery and data mining (<em>KDD 2021</em>). The paper is available <a href="https://arxiv.org/abs/2106.03819">here</a>.</p> <p>These datasets are used in the GitHub repository <a href="https://github.com/deezer/semi_perso_user_cold_start">deezer/semi_perso_user_cold_start</a> to reproduce experiments from the article.</p> <p>Please cite our paper if you use our code or data in your work.</p>
EMOPIA: A Multi-Modal Pop Piano Dataset For Emotion Recognition and Emotion-based Music Generation
<p>EMOPIA (pronounced ‘yee-mò-pi-uh’) dataset is a shared multi-modal (audio and MIDI) database focusing on perceived emotion in <strong>pop piano music</strong>, to facilitate research on various tasks related to music emotion. The dataset contains <strong>1,087</strong> music clips from 387 songs and <strong>clip-level</strong> emotion labels annotated by four dedicated annotators. </p> <p>For more detailed information about the dataset, please refer to our paper: <a href="https://arxiv.org/abs/2108.01374"><strong>EMOPIA: A Multi-Modal Pop Piano Dataset For Emotion Recognition and Emotion-based Music Generation</strong></a>. </p> <p><strong>File Description</strong></p> <ul> <li><em><strong>midis/</strong></em>: midi clips transcribed using GiantMIDI. <ul> <li>Filename `Q1_xxxxxxx_2.mp3`: Q1 means this clip belongs to Q1 on the V-A space; xxxxxxx is the song ID on YouTube, and the `2` means this clip is the 2nd clip taken from the full song.</li> </ul> </li> <li><em><strong>metadata/</strong></em>: metadata from YouTube. (Got when crawling)</li> <li> <p><em><strong>songs_lists/</strong></em>: YouTube URLs of songs.</p> </li> <li> <p><em><strong>tagging_lists/</strong></em>: raw tagging result for each sample.</p> </li> <li> <p><em><strong>label.csv</strong></em>: metadata that records filename, 4Q label, and annotator.</p> </li> <li> <p><em><strong>metadata_by_song.csv</strong></em>: list all the clips by the song. Can be used to create the train/val/test splits to avoid the same song appear in both train and test.</p> </li> <li> <p><em><strong>scripts/prepare_split.ipynb:</strong></em> the script to create train/val/test splits and save them to csv files.</p> </li> </ul> <p>------</p> <p><strong>2.2 Update</strong></p> <ul> <li>Add tagging files in <em><strong>tagging_lists/</strong></em> that are missing in the previous version.</li> <li>Add <em><strong>timestamps.json</strong></em> for easier usage. It records all the timestamps in dict format. You can see <em><strong>scripts/load_timestamp.ipynb</strong></em> for the format example.</li> <li>Add <em><strong>scripts/timestamp2clip.py</strong></em>: After the raw audio are crawled and put in <em><strong>audios/raw</strong></em>, you can use this script to get audio clips. The script will read <em><strong>timestamps.json</strong></em> and use the timestamp to extract clips. The clips will be saved to <em><strong>audios/seg</strong> </em>folder.</li> <li>remove 7 midi files that were added by mistake, and also corrected the number in <em><strong>metadata_by_song.csv</strong></em>.</li> </ul> <p> </p> <p><strong>2.1 Update</strong></p> <p>Add one file and one folder:</p> <ul> <li><em><strong>key_mode_tempo.csv</strong></em>: key, mode, and tempo information extracted from files.</li> <li><strong><em>CP_events/</em></strong>: CP events used in our paper. Extracted using this <a href="https://github.com/YatingMusic/compound-word-transformer/blob/main/dataset/representations/uncond/cp/corpus2events.py">script</a>, and add the emotion event to the front.</li> </ul> <p>Modify one folder:</p> <ul> <li>The <strong><em>REMI_events/</em></strong> files in version 2.0 contain some information that is not related to the paper, so remove it.</li> </ul> <p> </p> <p><strong>2.0 Update</strong></p> <p>Add two new folders:</p> <ul> <li><strong><em>corpus/</em></strong>: processed data that following <a href="https://github.com/YatingMusic/compound-word-transformer/blob/main/dataset/Dataset.md">the preprocessing flow</a>. (Please notice that although we have <code>1078</code> clips in our dataset, we lost some clips during steps 1~4 of the flow, so the final number of clips in this <strong><code>corpus</code></strong> is <code>1052</code>, and that's the number we used for training the generative model.)</li> <li><strong><em>REMI_events/</em></strong>: REMI event for each midi file. They are generated using this <a href="https://github.com/YatingMusic/compound-word-transformer/blob/main/dataset/representations/uncond/remi/corpus2events.py">script</a>.</li> </ul> <p>-------- </p> <p> </p> <p> </p> <p> </p> <p><strong>Cite this dataset</strong></p> <pre><code>@inproceedings{{EMOPIA}, author = {Hung, Hsiao-Tzu and Ching, Joann and Doh, Seungheon and Kim, Nabin and Nam, Juhan and Yang, Yi-Hsuan}, title = {{MOPIA}: A Multi-Modal Pop Piano Dataset For Emotion Recognition and Emotion-based Music Generation}, booktitle = {Proc. Int. Society for Music Information Retrieval Conf.}, year = {2021} }</code></pre>
MUHSIC: An Open Dataset with Temporal Musical Success Information
<p>Music is a volatile industry, where its dynamic nature can directly influence artist career behavior. That is, musical careers can suffer ups and downs depending on the current market moment. This dataset provides data about hot streak periods in musical careers, which are defined by high-impact bursts occurring in sequence.</p> <p>Success in the music industry has a temporal structure, as the audience tastes change over time. Here, we use the Billboard Hot 100 charts with Spotify data to represent success over time. For musical careers, we build their time series from the debut date (i.e., date of the first release obtained from Spotify) to the last chart collected. Thus, each point in the time series represents the success of such an artist in a given week, according to the Hot 100 chart. </p> <p>Therefore, we present <strong>MUHSIC</strong> (<strong>Mu</strong>sic-oriented <strong>H</strong>ot <strong>S</strong>treak <strong>I</strong>nformation <strong>C</strong>ollection), which contains:</p> <ul> <li><strong>Charts:</strong> enhanced data on all weekly Hot 100 Charts</li> <li><strong>Artists:</strong> artist success time series with hot streak information</li> <li><strong>Genres:</strong> genre success time series with hot streak information (the genre is the aggregated of all its artists)</li> <li><strong>Hot Streaks:</strong> summarized hot streak information</li> </ul>
jazznet: A Dataset of Fundamental Piano Patterns for Music Audio Machine Learning Research
<p>Jazznet is a dataset of piano patterns for music audio machine learning research. The dataset comprises chords, arpeggios, scales, and chord progressions in all keys of an 88-key piano and in all the inversions, for a total of 162520 labeled piano patterns, resulting in 95GB of data and more than 26k hours of audio. The data is also accompanied by Python scripts to enable the easy generation of new piano patterns beyond those present in the dataset. The data is broken down into small, medium, and large subsets, comprising 21516, 30328, and 52360 patterns, respectively (with all the chords, arpeggios, and scales being present in all subsets). </p> <p>The GitHub page of the dataset, containing details of the dataset and scripts for generating new data is https://github.com/tosiron/jazznet.</p>
Multitask Carnatic Music Dataset
<p>This dataset contains the multitask annotation for the Ragas in the Carnatic style of Indian classical music. The multitasks present in the dataset are the Swaras set, Melakarta set, Aaroh set, Avroh set, Janak/janya, and Raag Id. The music for extracting the mel-spectrogram feature is taken from the Dunya corpus [1]. The dataset contains 40 Ragas, and each contains 12 music recordings. </p> <p>[1] Porter, Alastair, Mohamed Sordo, and Xavier Serra. "Dunya: A system for browsing audio music collections exploiting cultural context." Britto A, Gouyon F, Dixon S. 14th International Society for Music Information Retrieval Conference (ISMIR); 2013 Nov 4-8; Curitiba, Brazil.[place unknown]: ISMIR; 2013. p. 101-6.. International Society for Music Information Retrieval (ISMIR), 2013.</p> <p> </p> <p>_________________________________________________________________________________________________________<br> This project was funded under the grant number: ECR/2018/000204 by the Science & Engineering Research Board (SERB).</p>
Multitask Hindustani Music Dataset
<p>This dataset contains the multitask annotation for the Raags in the Carnatic style of Indian classical music. The multitasks present in the dataset are the Swaras set, Jati, Thaat, Vadi, Samvadi, Aaroh set, Avroh set, and Raag Id. The music for extracting the mel-spectrogram feature is taken from the Dunya corpus [1]. The dataset contains 30 Ragas, and each contains 10 music recordings. </p> <p>[1] Porter, Alastair, Mohamed Sordo, and Xavier Serra. "Dunya: A system for browsing audio music collections exploiting cultural context." Britto A, Gouyon F, Dixon S. 14th International Society for Music Information Retrieval Conference (ISMIR); 2013 Nov 4-8; Curitiba, Brazil.[place unknown]: ISMIR; 2013. p. 101-6. International Society for Music Information Retrieval (ISMIR), 2013.</p> <p> </p> <p>_________________________________________________________________________________________________________<br> This project was funded under grant number: ECR/2018/000204 by the Science & Engineering Research Board (SERB).</p>
Hindustani Classical Music Transcription Dataset
<p>This dataset includes the transcriptions of Hindustani classical music recordings. Overall there are 430 pieces each having a 25-sec duration. These pieces include 329 Alap, 79 mid, and 22 end sections taken from 50 music recordings. Each transcription consists of annotations for Shrutis (22 in one octave) in 3 Shaptak (22 * 3 = 66), Alankar, and Silence or noise. We also include a symbol table mapping each annotation to a unique Unicode character.</p> <p> </p> <p>_________________________________________________________________________________________________________<br> This project was funded under grant number: ECR/2018/000204 by the Science & Engineering Research Board (SERB).</p>
DATASET OF THE "EL MUSICAL" THEATER ACOUSTIC PARAMETERS MEASURED ON-SITE
<p>Database of the acoustic parameters measured in the El Musical theater located in Valencia (Spain). There was one sound source position located in the middle of the proscenium and sixteen positions of the microphone in the room. Measurements were done twice, the first one with the stage opening curtain opened (points from 1 to 16) and the second one with it closed (points from 17 to 32). The database can be used to analyze the acoustic behavior of the theater, and to do an acoustic model of the room.</p>
Gamelan Music Dataset
<p><em>Gamelan</em> music is a traditional music from Indonesia. It is a handcrafted music ensemble. It is tuned based on the feeling of expert hearing called <em>Pangrawit</em>, hence, instrument frequencies among <em>Gamelan</em> ensembles are slightly different. Therefore, this dataset presents three different<em> Gamelan</em> music ensembles to serve more varied data. Audio from the first ensemble was collected from the Association of Music and Art at Institut Teknologi Bandung. Audio are isolated tone recordings from <em>Saron barung (Saron), Demung, Peking, Bonang barung, Bonang Penerus, Slenthem, </em>and <em>Kendhang</em> instruments. A <em>Gamelan</em> expert arranged orchestra recordings from isolated tone recordings using FL studio software. The targets of the instrument sources in the orchestra are also provided. Furthermore, Audio from the second ensemble was obtained from the GamelanTron Project at Universitas Dian Nuswantoro. GamelanTron project acquired audio recordings from Kendhang rhythms to control the tempo of the other <em>Gamelan</em> instruments. The last ensemble was used <em>Gamelan</em> ensemble from Institut Teknologi Sepuluh Nopember. The audio is isolated tone recordings from <em>Saron barung, Peking, </em>and <em>Demung </em>instruments. The number of audio in the<em> Gamelan</em> music dataset is 279 recordings with various durations. This dataset can be used in music information retrieval research.</p> <p> </p>
Dataset for "Learning to Read and Follow Music in Complete Score Sheet Images"
<p>This is the data used in the paper:<br> Florian Henkel , Rainer Kelz, Gerhard Widmer. "Learning to Read and Follow Music in Complete Score Sheet Images". Proceedings of the 21st International Society for Music Information Retrieval Conference (ISMIR), 2020</p> <p>The dataset is based on the <a href="https://zenodo.org/record/2597505#.XxbJtxGxVhE">MSMD - Multimodal Sheet Music Dataset</a> and processed for the purpose of score following in complete score sheet images.</p> <pre>Please find further information and the corresponding code on this Github page: <a href="https://github.com/CPJKU/audio_conditioned_unet">https://github.com/CPJKU/audio_conditioned_unet</a> </pre>
Dataset for the paper "The skipping behavior of users of music streaming services and its relation to musical structure"
<p>Dataset for the paper "The skipping behavior of users of music streaming services and its relation to musical structure" by Nicola Montecchio, Pierre Roy, François Pachet - 10.1371/journal.pone.0239418</p>
Dataset from the ISMIR2020 article "Multilingual Music Genre Embeddings for Effective Cross-Lingual Music Item Annotation"
<p>We release the data required to reproduce the cross-lingual music genre translation experiments from the article <strong><em>Multilingual Music Genre Embeddings for Effective Cross-Lingual Music Item Annotation</em></strong> presented at the <a href="https://ismir.github.io/ISMIR2020/">ISMIR 2020</a> conference.</p> <p>More information about this data and how it should be used in the experiments can be found in the GitHub repository <a href="http://github.com/deezer/MultilingualMusicGenreEmbedding">deezer/MultilingualMusicGenreEmbedding</a>.</p> <p>Please cite our paper if you use the code or data in your work.</p>
GUITAR-FX-DIST: A Dataset of Processed Guitar Recordings for Music Research - (Mono Continuous)
<p><strong>GUITAR-FX-DIST</strong> is a dataset of electric guitar recordings processed with overdrive, distortion and fuzz audio effects. It was developed for research in guitar effects detection, classification and parameters estimation. The dataset is also useful for research on automatic music transcription, intelligent music production, signal processing or effects modelling. It contains both unprocessed and processed recordings.</p> <p>The dataset is split into 4 sub-datasets: Mono Continuous, Mono Discrete, Poly Continuous, Poly Discrete</p> <p> </p> <p><strong>Authors:</strong></p> <p>Marco Comunità - <a href="http://c4dm.eecs.qmul.ac.uk/">Centre for Digital Music</a>, Queen Mary University of London</p> <p> </p> <p><strong>Reference:</strong></p> <p>If you make use of GUITAR-FX-DIST, please cite the following publication:</p> <pre><code>@article{comunità2021guitar, title={Guitar Effects Recognition and Parameter Estimation with Convolutional Neural Networks}, author={Comunità, Marco and Stowell, Dan and Reiss, Joshua D.}, journal={Journal of the Audio Engineering Society}, year={2021}, volume={69}, number={7/8}, pages={594-604}, doi={}, month={July} }</code></pre> <p> </p> <p><strong>Dataset Snapshot:</strong></p> <ul> <li><strong>Size:</strong> ~550k samples (~305 hours) + 550k mel spectrograms</li> <li><strong>Audio Format:</strong> WAV - 44.1kHz, 16bit, mono, -6dBFS</li> <li><strong>Mel-Spectrogram Format:</strong> NPY - 128 frequency bands, sample rate 22050Hz, window length 1024, hop size 512,</li> <li><strong>Effects:</strong> 14 between overdrive, distortion and fuzz</li> <li><strong>Unprocessed recordings</strong> <ul> <li>624 monophonic notes</li> <li>420 polyphonic (2, 3 and 4 notes intervals and chords)</li> <li>2 guitars, with up to 2 pick-up settings and up to 3 plucking styles (finger pluck - hard, finger pluck - soft, pick) <ul> <li>Schecter Diamond C-1 Classic</li> <li>Chester Stratocaster</li> </ul> </li> </ul> </li> <li><strong>Samples length:</strong> 2 sec</li> </ul> <p> </p> <p><strong>Unprocessed Recordings:</strong></p> <p>The original (unprocessed) recordings are from the <a href="https://www.idmt.fraunhofer.de/en/business_units/m2d/smt/audio_effects.html">IDMT-SMT-Audio-Effects</a> dataset.</p> <p>For details please refer to the website and the accompagning publication:</p> <p><em>Stein, Michael; Abeßer, Jakob; Dittmar, Christian; Schuller, Gerald: Automatic Detection of Audio Effects in Guitar and Bass Recordings. Proceedings of the AES 128th Convention, 2010.</em></p> <p> </p> <p><strong>Processed Recordings:</strong></p> <p>The processed recordings are divided into 4 sub-datasets which are named depending on the unprocessed recordings used (monophonic or polyphonic) and on the settings' values (discrete or continuous).</p> <p>The sub-datasets are called: Mono Discrete, Poly Discrete, Mono Continuous, Poly Continuous</p> <p>Mono Discrete and Poly Discrete use a discrete set of combinations selected as the most common and representative settings a person might use (see README file for details).</p> <p>For Mono Continuous and Poly Continuous both unprocessed samples as well as settings’ values are drawn from a uniform distribution (10000 samples for each effect).</p> <p>Samples:</p> <ul> <li>Mono Discrete: ~160k</li> <li>Poly Discrete: ~110k</li> <li>Mono Continuous: 140k</li> <li>Poly Continuous: 140k</li> </ul> <p> </p> <p><strong>Scripts:</strong></p> <p>The dataset includes the MATLAB scripts used to generate the samples</p>
HHDS - Spanish HipHop Dataset for Music Source Separation
<p><strong>What is HHDS?</strong></p> <p>HHDS is a reduced compilation of Hip Hop songs, used to train a Convolutional Neural Network (CNN) for audio source separation in [1], built on top of the DeepConvSep framework [2] developed at the Music Technology Group (MTG), Universitat Pompeu Fabra.</p> <p>The structure of HHDS follows the convention of DSD100 [3] (Demixing Secrets Dataset). HHDS contains the separated tracks for the categories of bass, drums, vocals and others in monophonic WAV les with a sampling rate of 44100Hz. The mixture is calculated by normalizing the sum of the tracks. The main difference with respect to DSD100 is that in HHDS there are HipHop songs only, instead of many different genres. The total number of songs is 18, from which 13 are used for training and 5 are used for evaluation.</p> <p>A detailed list of the songs included in the dataset can be found inside the .zip file provided. The reader can also find the code for this dataset in the DeepConvSep repository in the path examples/hiphopss.</p> <p> </p> <p><strong>References</strong></p> <p>[1] "A Deep Learning Approach to Source Separation and Remixing of HipHop music", Héctor Martel, Undergraduate Thesis, Universitat Pompeu Fabra 2016-2017. </p> <p>[2] DeepConvSep repository on GitHub: https://github.com/MTG/DeepConvSep</p> <p>[3] Demixing Secrets Dataset (DSD100), SiSEC2016: http://liutkus.net/DSD100.zip</p>
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.