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.
41
datasets available to search
ShareScore release 0.9.0
Dataset results
41 results for “spectrogram”
Detecting repeating earthquakes on the San Andreas Fault with unsupervised machine-learning of spectrograms (supplementary material)
<p>Supplementary material for Sawi et al., 2023, <i>Detecting repeating earthquakes on the San Andreas Fault with unsupervised machine-learning of spectrograms </i>(The Seismic Record). Catalog of repeating earthquakes in sequences on a 10-km long segment of the San Andreas Fault in California from 1984-2019. </p><p> </p><p><strong>Catalog Header</strong></p><p>YR/MO/DY...........Date of event</p><p>HR/MN/SC...........Time of event</p><p>LAT/LON/DEP........Location of event</p><p>EX/EY/EZ...........Relative location uncertainty (in m)</p><p>MAG................NCSN magnitude</p><p>evID.................NCSN event ID</p><p>seqID................Repeating earthquake sequence ID</p><p>isRESp............Is quasi-periodic RES (bool)</p><p> </p><p><strong>References: </strong></p><p>Sawi T., Waldhauser F., Holtzman B. K., Groebner, N. (2023) Detecting repeating earthquakes on the San Andreas Fault with unsupervised machine-learning of spectrograms. The Seismic Record. </p><p>Waldhauser, F., and Schaff, D. P. (2021). A Comprehensive Search for Repeating Earthquakes in Northern California: Implications for Fault Creep, Slip Rates, Slip Partitioning, and Transient Stress. J Geophys Res B Solid Earth, 126(11), 1–22. <a href="https://doi.org/10.1029/2021JB022495">https://doi.org/10.1029/2021JB022495</a></p>
Image-based Classification of Intense Radio Bursts from Spectrograms: An Application to Saturn Kilometric Radiation
<p>A catalogue of 4874 of the Low Frequency Extensions (LFEs) of Saturn Kilometric Radiation (SKR) detected by Cassini/RPWS from the beginning of 2004 until mission end in 2017. The LFEs presented in this catalogue were identified using a modified U-Net architecture that applied semantic segmentation to spectrogram images in order to extract the exact frequency-time coordinates of the LFE. The files consist of a .json file with the coordinates of each LFE in Time Frequency Catalogue (TFCat) format (Cecconi et. al. 2023). We also include a .csv file with the start and stop times of each LFE in the form of python datetime timestamps, with the average predicted probability per LFE as an accompanying column. </p>
Catalogue of Arctic Toothed Whale Click Spectra and Spectrograms from SoundTrap Recordings
<p>This supplementary file accompanies a research article focused on beluga and narwhal echolocation click classification. This document provides an overview of the data included in all beluga and narwhal acoustic events (each 1 hour) from SoundTrap recordings collected at the Kong Oscar and Fisher Islands mooring sites. There are two figures for each event: (1) concatenated click spectrogram, and (2) mean power spectrum for all detections within the event. Summary statistics are provided for each event showing the number of detections and the one-third octave level (TOL) ratio in decibels (dB) between selected TOL bands. Visual inspection of the enclosed figures demonstrates clear differences between beluga and narwhal click spectra.</p>
Spectrograms and frequencies of the first mode of Schumann Resonance according to multi-position monitoring parformed at the Ukrainian Antarctic Station (2002-2020) and at the Arctic SOUSY facility (2013-2020)
<p>This dataset contains the processed data used for the publication: ELECTROMAGNETIC SEASONS IN SCHUMANN RESONANCE RECORDS. The dataset contains the daily spectrograms and frequensies of first mode of Schumann Resonance (derived for North-South and East-West magnetic components) of ELF signals recorded at the Ukrainian “Akademik Vernadsky” Antarctic station (65.25° N and 64.25° W) 2002-2020, and at SOUSY Arctic facility (Svalbard 78.15° N and 16.05° E) 2013-2020.</p>
Fig. 3. Spectrograms and oscillograms for Dendropsophus vraemi. A–B in The distribution and calls of Vraem' Treefrog, Dendropsophus vraemi (Caminer, Milá, Jansen, Fouquet, Venegas, Chávez, Lougheed, and Ron 2017), with comments on its conservation status
Fig. 3. Spectrograms and oscillograms for Dendropsophus vraemi. A–B: variations of the advertisement call; C: an aggressive call. A, B and C are call variations from a single male individual (CORBIDI 17894) recorded in the middle Apurimac basin.
Ambient sound spectrograms between 2015-01-01 and 2021-01-01 for OOI low-frequency hydrophones
<p>Ambient sound spectrograms calculated for the OOI low-frequency hydrophones. The spectrograms are PSD estimates using the welch method and median averaging with an averaging time of 15 minutes and 512 points per segment.</p>
BRAMS Radio Spectrograms and Spectrogram Samples for Automatic Detection of Meteor Echoes
<p>The files in this dataset are based of radio recordings taped by BRAMS (Belgian RAdio Meteor Stations), the Belgian meteor detection network.</p> <p>Included in the dataset are the original BRAMS radio recordings (stored as .wav audio files), the spectrogram data for each radio recording (stored as .csv files) and the meteor and non-meteor samples extracted from the radio spectrograms (stored as .csv files).</p> <p>It should be noted that the the spectrogram data was sampled using a sliding window of size 30x20 pixels and the samples extracted in this manner were further processed by calculating the vertical average of each column in the 30x20 matrixes. The result of this sampling procedure is a set of data vectors containing the average power of the signal found in the original 30x20 spectrogram sample.</p>
Data for spectrogram and waveforms
<p>This dataset contains the ascii data for spectrogram and waveforms observed by distributed acoustic sensing with the Muroto cable.</p>
Supplementary material S17: All instances of jolting pulses on honeycomb where the vibrational trace is clearly visible (spectrograms).
<p>A series of spectrograms demonstrating the most clearly visible <em>Varroa </em>jolting vibrational pulses registered on honeycomb. These spectrograms showcase the larger variation that is observed in jolting pulses on this substrate. Panels e, f and h provide evidence for the broad-band and generation of signal at the high-frequency bandwidth. The magnitude of acceleration is logarithmic (to the base 10), where the maximum is in red (1x10<sup>-3</sup> m/s<sup>2</sup>) and the minimum blue (and forced to be 1/20 of the maximum).</p>
Data from: Spectrogram cross-correlation can be used to measure the complexity of bird vocalizations
<p>This data set contains catalog numbers of Macaulay Library sound files and RVV library sound files used in the study- Sawant, S; Arvind, C; Joshi, V & Robin, V. V. (2021) Spectrogram cross-correlation can be used to measure the complexity of bird vocalizations.</p>
Mode Splitting Spectrogram and Analysis Plots, 11 May 2022
<p>Spectrogram collected in Spectrum Lab by Steve WA5FRF. Digitized manually using MATLAB. Digitization and computation code at <a href="https://github.com/KCollins/wa5frf-plots">https://github.com/KCollins/wa5frf-plots</a>.</p>
Spectrograms of Electric Guitar Notes in the SemanticTimbreDataset
<p>This dataset contains spectrograms for all monophonic electric guitar notes contained within the SemanticTimbreDataset.</p> <p>Files are organised as follows: Timbre_Group/Timbre_Descriptor/Timbre_Magnitude.</p> <p>All audio files with a timbre magnitude of 0 are monophonic electric guitar notes recorded from a Fender Stratocaster, originally sourced from the EGFxSet [1]. All other audio files are those original clean samples processed through various guitar pedals affecting the sounds at various intensities (timbre magnitude). Each guitar pedal's effect can be described by a separate timbre descriptor. The SemanticTimbreDataset contains 19 timbre descriptors in total.</p> <p> </p> <p>[1] Hegel Pedroza, Gerardo Meza, & Iran R. Roman. (2022). EGFxSet: Electric guitar tones processed through real effects of distortion, modulation, delay and reverb (Version 1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.7044411</p>
Spectrogram of vibrations from inside an active beehive (Apis mellifera)
<p>A piezoelectric transducer is inserted inside an active beehive. We record the vibrations within the hive every 5 minutes for about 2 months. We extract the spectrogram from each recording. The spectrogram is a time-frequency representation that shows how the frequencies of the recording evolve over time. All spectrograms from 3865 recordings are stacked and their corresponding timestamp appears on the upper left corner. The video demonstrates the vibrational activity of the bees and any other factor that can vibrate the beehive for two months.</p>
Spectrograms of the SOL dataset
<p>This is the dataset accompanying the paper <a href="https://arxiv.org/abs/1906.08152">Learning Disentangled Representations of Timbre and Pitch for Musical Instrument Sounds Using Gaussian Mixture Variational Autoencoders</a> published at ISMIR2019.</p> <p>Due to the copyright, only the extracted spectrograms are shared.</p> <p>The repo is at https://github.com/yjlolo/gmvae-synth.</p>
FIGURE 4. Calling song spectrograms. A in Two new species of Hygronemobius Hebard, 1913 (Orthoptera, Grylloidea, Nemobiinae) from Brazilian Amazon
FIGURE 4. Calling song spectrograms. A—three chirps of Hygronemobius duckensis sp. nov., the first with 20 pulses and the other ones with 21 pulses; B—nine chirps of Hygronemobius dialeucus sp. nov., first eight with five pulses and the last one with four pulses.
FIGURE 5.—Spectrograms. A in Anurogryllus Saussure, 1877 (Orthoptera: Gryllidae: Gryllinae) from southern Brazil: new species and new records
FIGURE 5.—Spectrograms. A—Anurogryllus tapes sp. nov.; B—Anurogryllus patos sp. nov.; C—Anurogryllus tolepizai (de Mello, 1988).
Supplementary material S16: All honeycomb jolting pulses shown as individual spectrograms.
<p>Video showcasing all 28 honeycomb jolting pulse spectrograms. All pulses are showcased in the same way as those seen in S12 and S14 and share the same analysis. The magnitude of acceleration is logarithmic (to the base 10), where the highest magnitude is 6.5x10<sup>-4 </sup>m/s<sup>2</sup>, and the lowest magnitude set to be 1/40 of the maximum.</p>
Supplementary material S12: All petri-dish jolting pulses shown as individual spectrograms.
<p>Video showcasing the full collection of 250 petri-dish <em>Varroa</em> jolting pulses. Every pulse in the collection is aligned to the centre of the window and presented in decreasing order of strength. The full breadth of variation can be seen between the jolting pulses when viewing them in this way. The magnitude of acceleration is logarithmic (to the base 10) where the highest magnitude is 2x10<sup>-3 </sup>m/s<sup>2</sup> and the lowest magnitude forced to be 1/40 of the maximum to reduce the contribution of meaningless noise.</p>
Supplementary material S14: All brood-comb jolting pulses shown as individual spectrograms.
<p>Video showcasing all 189 brood-comb jolting pulse spectrograms. The jolting pulse spectra have undergone the same centring and analysis as in S12 and S16. The magnitude of acceleration is logarithmic (to the base 10), where the highest magnitude is 2x10<sup>-3 </sup>m/s<sup>2</sup>, and the lowest magnitude set to be 1/40 of the maximum.</p>
Beat This! Spectrograms for Beat and Downbeat Tracking
<p>This collection contains mel spectrograms and annotations of 16 datasets for beat and downbeat tracking. All datasets have been used in "<a href="https://arxiv.org/abs/2407.21658">Beat This! Accurate beat tracking without DBN postprocessing</a>" (Foscarin/Schlüter/Widmer, ISMIR 2024) and prior publications by other authors, but for many of these datasets, audio data is not publicly available. By publishing the spectrograms, we invite other researchers to improve the state of the art in beat and downbeat tracking.</p> <h3>Datasets</h3> <p>Spectrograms for the following datasets are included in the collection:</p> <ul> <li><a href="https://github.com/fosfrancesco/asap-dataset">asap</a>: "ASAP: a dataset of aligned scores and performances for piano transcription" (Foscarin et al., ISMIR 2020)</li> <li>ballroom: "An experimental comparison of audio tempo induction algorithms" (Gouyon et al., TASLP 2006) for the audio and "Rhythmic Pattern Modeling for Beat and Downbeat Tracking in Musical Audio" (Krebs/Böck/Widmer, ISMIR 2013) for the annotations</li> <li>beatles: "Evaluation methods for musical audio beat tracking algorithms" (Davies/Degara/Plumbley, Tech. Rep., QMU, 2019)</li> <li><a href="https://www.eumus.edu.uy/candombe/datasets/ISMIR2015/dataset.html">candombe</a>: "Beat and Downbeat Tracking Based on Rhythmic Patterns Applied to the Uruguayan Candombe Drumming" (Nunes et al., ISMIR 2015) </li> <li>filosax: "Filosax: A dataset of annotated jazz saxophone recordings" (Foster/Dixon, ISMIR 2021)</li> <li><a href="https://magenta.tensorflow.org/datasets/groove" rel="nofollow">groove_midi</a>: "Learning to groove with inverse sequence transformations" (Gillick et al., ICML 2019)</li> <li>gtzan: "Musical genre classification of audio signals" (Tzanetakis/Cook, TSAP 2002) for the audio and "Swing ratio estimation" (Marchand/Peters, DAFx 2015) for the annotations</li> <li><a href="https://zenodo.org/records/3371780">guitarset</a>: "GuitarSet: A dataset for guitar transcription" (Xi et al., ISMIR 2018)</li> <li>hainsworth: "Particle filtering applied to musical tempo tracking" (Hainsworth/Macleod, JASP 2004)</li> <li>harmonix: "The Harmonix set: Beats, downbeats, and functional segment annotations of western popular music" (Nieto et al., ISMIR 2019) for the original and "Modeling Beats and Downbeats with a Time-Frequency Transformer" (Hung et al., ICASSP 2022) for the version included here</li> <li>hjdb: "One in the jungle: Downbeat detection in hardcore, jungle, and drum and bass" (Hockman/Davies/Fujinaga, ISMIR 2012)</li> <li>jaah: "Audio-aligned jazz harmony dataset for automatic chord transcription and corpus-based research" (Eremenko et al., ISMIR 2018)</li> <li><a href="https://staff.aist.go.jp/m.goto/RWC-MDB/" rel="nofollow">rwc</a>: "RWC music database: Popular, classical and jazz music databases" (Goto et al., ISMIR 2002) for the audio and "AIST annotation for the RWC music<br>database" (Goto, ISMIR 2006) for the annotations</li> <li>simac: "A computational approach to rhythm description — Audio features for the computation of rhythm periodicity functions and their use in tempo induction and music content processing" (Gouyon, PhD thesis, UPF, 2005)</li> <li><a href="https://joserzapata.github.io/publication/selective-sampling-beat-tracking/">smc</a>: "Selective sampling for beat tracking evaluation" (Holzapfel et al., TASLP 2012)</li> <li>tapcorrect: "Towards Automatically Correcting Tapped Beat Annotations for Music Recordings" (Driedger et al., ISMIR 2019)</li> </ul> <p>If given, links in the above list point to locations for obtaining the original audio.</p> <h3>Annotations</h3> <p>The corresponding annotations are available on <a href="https://github.com/CPJKU/beat_this_annotations">https://github.com/CPJKU/beat_this_annotations</a>. A snapshot of v1.0 is included in this collection as <code>beat_this_annotations.zip</code>, but you may want to use a later release.</p> <h3>Spectrograms</h3> <p>Spectrograms are computed from monophonic audio at a sample rate of 22050 Hz with a window size of 1024 and hop size of 441 samples (yielding 50 frames per second), processed with a mel filterbank of 128 bands from 30 Hz to 11 kHz, and magnitudes scaled with ln(1+1000x). They are provided in half-precision floating-point format. Spectrograms can be reproduced with torchaudio 2.3.1 from a 22050 Hz waveform tensor (resampled with <code><a href="https://python-soxr.readthedocs.io/en/latest/soxr.html#soxr.resample">soxr.resample()</a></code>, if needed) via:</p> <pre><code>melspect = torchaudio.transforms.MelSpectrogram(sample_rate=22050, n_fft=1024, hop_length=441, f_min=30, f_max=11000, n_mels=128, mel_scale='slaney', normalized='frame_length', power=1)(waveform).mul(1000).log1p()</code></pre> <h3>Format</h3> <p>For each dataset, a compressed .zip file is provided, which in turn holds an uncompressed .npz file. The .npz file holds a set of numpy arrays in subdirectories named after the annotations. Each subdirectory contains a spectrogram of the original audio file ("track.npy"), 11 pitch-shifted versions from -5 to +6 semitones ("track_ps-5.npy" to "track_ps6.npy") and 10 time-stretched versions from -20% to +20% ("track_ts-20.npy" to "track_ts20.npy"), except for <code>gtzan.npz</code>, which is designated for testing and only holds the original audio files. The .npz files can be loaded in numpy via <code><a href="https://numpy.org/doc/2.0/reference/generated/numpy.load.html">np.load()</a></code>, or unzipped into a set of .npy files that can again be loaded via <a href="https://numpy.org/doc/2.0/reference/generated/numpy.load.html"><code>np.load()</code></a>. We also provide <a href="https://github.com/CPJKU/beat_this/blob/main/beat_this/dataset/mmnpz.py">code to load .npz files as memory maps</a> for more efficiency.</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.