Summary
MulTTiPop is a new dataset for evaluating automatic music transcription (AMT) models, presented in the arXiv paper 2507.08753 by Nathan Pruyne, Benjamin Stoler, and William Chen (July 2025). It pairs 572 pop music segments (3.5 hours of audio) with multitrack MIDI recordings, covering diverse genres and decades from the 1930s to the 2000s. The dataset was built by performing metadata-based matching on song segments from the Lakh MIDI and TheoryTab datasets, manually identifying an anchor beat between audio and MIDI, then applying beat tracking to the audio and warping the MIDI to match its tempo and timing. Benchmarking state-of-the-art AMT models on MulTTiPop revealed substantial room for improvement, with the best model achieving only 38% Onset F1 score, indicating that pop music transcription remains a challenging open problem.
Overview
Research area: Music AI
Authors: Nathan Pruyne, Benjamin Stoler, William Chen
Published: 2025-07-12
arXiv: 2507.08753
Abstract
We present MulTTiPop, a dataset of pop music segments and their associated multitrack MIDI recordings for the evaluation of automatic music transcription models. MulTTiPop contains 572 segments of popular music totaling 3.5 hours of audio, and contains songs from diverse genres and decades from the 1930s to 2000s.
To collect this dataset, the authors:
- Perform metadata-based matching on song segments from the Lakh MIDI and TheoryTab datasets
- Manually identify an anchor beat between the audio and MIDI
- Use beat tracking on the audio and warp the MIDI to match its tempo and timing
They evaluate state-of-the-art automatic music transcription models on MulTTiPop and find substantial room for improvement, with the best model achieving
38% Onset F1, showing that transcription of real-world pop music remains a challenging benchmark.
More details and sound examples of MulTTiPop are available via the paper: https://arxiv.org/abs/2507.08753
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/178379426