# Sound and Music Biases in Deep Music Transcription Models: A Systematic Analysis

arXiv: https://arxiv.org/abs/2512.14602 · HTML: https://mlsanity.com/paper/2512.14602/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 37 references that are in the map.
- First posted on arXiv 16 Dec 2025.

## Builds on (most cited first, 16 of 16)

- [Enabling Factorized Piano Music Modeling and Generation with the MAESTRO Dataset](https://mlsanity.com/paper/1810.12247/) (2018)
- [Onsets and Frames: Dual-Objective Piano Transcription](https://mlsanity.com/paper/1710.11153/) (2017)
- [High-resolution Piano Transcription with Pedals by Regressing Onset and Offset Times](https://mlsanity.com/paper/2010.01815/) (2020)
- [Sequence-to-Sequence Piano Transcription with Transformers](https://mlsanity.com/paper/2107.09142/) (2021)
- [ID and OOD Performance Are Sometimes Inversely Correlated on Real-world Datasets](https://mlsanity.com/paper/2209.00613/) (2022)
- [Foundation Models for Music: A Survey](https://mlsanity.com/paper/2408.14340/) (2024)
- [LLark: A Multimodal Instruction-Following Language Model for Music](https://mlsanity.com/paper/2310.07160/) (2023)
- [Computer-Generated Music for Tabletop Role-Playing Games](https://mlsanity.com/paper/2008.07009/) (2020)
- [Automatic Piano Transcription with Hierarchical Frequency-Time Transformer](https://mlsanity.com/paper/2307.04305/) (2023)
- [On the Potential of Simple Framewise Approaches to Piano Transcription](https://mlsanity.com/paper/1612.05153/) (2016)
- [A Data-Driven Analysis of Robust Automatic Piano Transcription](https://mlsanity.com/paper/2402.01424/) (2024)
- [Polyphonic Piano Transcription Using Autoregressive Multi-State Note Model](https://mlsanity.com/paper/2010.01104/) (2020)
- [Towards Musically Informed Evaluation of Piano Transcription Models](https://mlsanity.com/paper/2406.08454/) (2024)
- [An Experimental Analysis of the Entanglement Problem in Neural-Network-based Music Transcription Systems](https://mlsanity.com/paper/1702.00025/) (2017)
- [A Computational Study of the Role of Tonal Tension in Expressive Piano Performance](https://mlsanity.com/paper/1807.01080/) (2018)
- [Piano Transcription by Hierarchical Language Modeling with Pretrained Roll-based Encoders](https://mlsanity.com/paper/2501.03038/) (2025)

## Built on by (most cited first, 2 of 2)

- [Exploring LLMs for South Asian Music Understanding and Generation](https://mlsanity.com/paper/2606.05522/) (2026)
- [Learning to Predict Performance-induced Emotion Differences in Classical Piano Music](https://mlsanity.com/paper/2607.28876/) (2026)

## Abstract (arXiv)

Automatic Music Transcription (AMT) -- the task of converting music audio into note representations -- has seen rapid progress, driven largely by deep learning systems. Due to the limited availability of richly annotated music datasets, much of the progress in AMT has been concentrated on classical piano music, and even a few very specific datasets. Whether these systems can generalize effectively to other musical contexts remains an open question. Complementing recent studies on distribution shifts in sound (e.g., recording conditions), in this work we investigate the musical dimension -- specifically, variations in genre, dynamics, and polyphony levels. To this end, we introduce the MDS corpus, comprising three distinct subsets -- (1) Genre, (2) Random, and (3) MAEtest -- to emulate different axes of distribution shift. We evaluate the performance of several state-of-the-art AMT systems on the MDS corpus using both traditional information-retrieval and musically-informed performance metrics. Our extensive evaluation isolates and exposes varying degrees of performance degradation under specific distribution shifts. In particular, we measure a note-level F1 performance drop of 20 percentage points due to sound, and 14 due to genre. Generally, we find that dynamics estimation proves more vulnerable to musical variation than onset prediction. Musically informed evaluation metrics, particularly those capturing harmonic structure, help identify potential contributing factors. Furthermore, experiments with randomly generated, non-musical sequences reveal clear limitations in system performance under extreme musical distribution shifts. Altogether, these findings offer new evidence of the persistent impact of the Corpus Bias problem in deep AMT systems.