This dataset was created for the paper:
MUSCAT: a Multimodal mUSic Collection for Automatic Transcription of real recordings and image scores
Alejandro Galan-Cuenca, Jose J. Valero-Mas, Juan C. Martínez-Sevilla, Antonio Hidalgo-Centeno, Antonio Pertusa, Jorge
Calvo-Zaragoza
Accepted for oral presentation in ACM Multimedia, 2024
About the dataset
MUSCAT is an assortment of acoustic recordings, image sheets, and their score-level annotations in several notation formats.
Despite a large number of existing works in the Automatic Music Transcription (AMT) field, there is a shortage of end-to-end Audio-to-Score (A2S) transcription efforts, leading to a lack of benchmark corpora, particularly when dealing with real data.
This dataset comprises almost 80 hours of real recordings with varied instrumentation and polyphony degrees (from piano to orchestral music), 1251 scanned sheets, and 880 symbolic scores from 37 composers, which may also be used in other tasks involving metadata such as instrument identification or composer recognition.
A fragmented subset of this collection exclusively focused on acoustic data for score-level AMT (MUSCUTS assortment) is also presented together with a baseline experimentation using short audio excerpts. Finally, a web-based service is also provided to allow increasing the size of the collections collaboratively.
Request the dataset download link. The dataset (in .tgz format) includes all the recordings, digital scores, and images. Optionally, you can also request access to the web interface to browse and contribute to the dataset.
-
Download the code for the end-to-end A2S model used in the paper as the baseline.
-
Download the MUSCUTS tools, which include the Musescore plugin to mark timestamps in audio files and the code to split them into short excerpts.
The dataset is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License.
This work is part of the I+D+i MultiScore project with code PID2020-118447RA-I00, funded by MCIN/AEI/10.13039/501100011033.
