arXiv:2607.08756v1 Announce Type: cross 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, we perform metadata-based matching on song segments from the Lakh MIDI and TheoryTab datasets, manually identify an anchor beat between the audio and MIDI, then use beat tracking on the audio and warp the MIDI
The continuous drive for more robust and diverse datasets is critical for advancing AI models in specialized domains like music transcription.
Improved datasets like MulTTiPop enable more accurate and versatile AI models, reducing the manual effort required for music analysis and content creation.
The availability of MulTTiPop expands the resources for training and evaluating automatic music transcription systems, potentially leading to better AI-driven music tools.
- · AI music startups
- · Music producers
- · Academic researchers
- · Music entertainment industry
- · Manual transcription services (long term)
Automatic music transcription models will become more accurate and capable of handling diverse popular music.
This improved accuracy will accelerate the development of AI tools for music creation, analysis, and educational applications.
The integration of advanced AI in music could democratize music production and analysis, allowing more individuals to create and understand music at a technical level.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG