ACPAS is a dataset with aligned audio and scores for classical piano music containing 497 distinct music scores aligned with 2189 performances, in total 179.77 hours. For each performance, we provide the corresponding performance audio (real recording or synthesized recording), performance MIDI, and MIDI score, together with rhythm and key annotations. This is the Real recording subset of the ACPAS dataset. To download the full dataset and for dataset details, please refer to the dataset webpage at https://cheriell.github.io/research/ACPAS_dataset For any questions, suggestions, or comments, please do not hesitate to contact lele.liu@qmul.ac.uk How to cite: - Lele Liu, Veronica Morfi, and Emmanouil Benetos, "ACPAS: A Dataset of Aligned Classical Piano Audio and Scores for Audio-to-Score Transcription," in ISMIR Late-breaking Demo, 2021. Funding: L. Liu is a research student at the UKRI Centre for Doctoral Training in Artificial Intelligence and Music, supported jointly by the China Scholarship Council and Queen Mary University of London.