End-to-end Piano Performance-MIDI to Score Conversion with Transformers
This work addresses a fundamental task in computational musicology for musicians and researchers, offering a novel approach that simplifies annotation and improves accuracy, though it is incremental in applying transformers to a specific domain.
The paper tackles the problem of automatically converting expressive piano performances into accurate musical notation by introducing an end-to-end transformer-based method that directly predicts detailed scores from MIDI files, achieving significant improvements in transcription metrics like MUSTER over previous approaches. It reduces sequence lengths by 3.5 times compared to prior methods and is the first to predict notational details such as trill marks from performance data.
The automated creation of accurate musical notation from an expressive human performance is a fundamental task in computational musicology. To this end, we present an end-to-end deep learning approach that constructs detailed musical scores directly from real-world piano performance-MIDI files. We introduce a modern transformer-based architecture with a novel tokenized representation for symbolic music data. Framing the task as sequence-to-sequence translation rather than note-wise classification reduces alignment requirements and annotation costs, while allowing the prediction of more concise and accurate notation. To serialize symbolic music data, we design a custom tokenization stage based on compound tokens that carefully quantizes continuous values. This technique preserves more score information while reducing sequence lengths by $3.5\times$ compared to prior approaches. Using the transformer backbone, our method demonstrates better understanding of note values, rhythmic structure, and details such as staff assignment. When evaluated end-to-end using transcription metrics such as MUSTER, we achieve significant improvements over previous deep learning approaches and complex HMM-based state-of-the-art pipelines. Our method is also the first to directly predict notational details like trill marks or stem direction from performance data. Code and models are available at https://github.com/TimFelixBeyer/MIDI2ScoreTransformer