# Unsupervised Lead Sheet Generation via Semantic Compression

arXiv: https://arxiv.org/abs/2310.10772 · HTML: https://mlsanity.com/paper/2310.10772/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 25 references that are in the map.
- First posted on arXiv 16 Oct 2023.

## Builds on (most cited first, 19 of 19)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [POP909: A Pop-song Dataset for Music Arrangement Generation](https://mlsanity.com/paper/2008.07142/) (2020)
- [Reparameterizable Subset Sampling via Continuous Relaxations](https://mlsanity.com/paper/1901.10517/) (2019)
- [A Probabilistic Formulation of Unsupervised Text Style Transfer](https://mlsanity.com/paper/2002.03912/) (2020)
- [The Jazz Transformer on the Front Line: Exploring the Shortcomings of AI-composed Music through Quantitative Measures](https://mlsanity.com/paper/2008.01307/) (2020)
- [Anticipatory Music Transformer](https://mlsanity.com/paper/2306.08620/) (2023)
- [Controllable deep melody generation via hierarchical music structure representation](https://mlsanity.com/paper/2109.00663/) (2021)
- [Multitrack Music Transformer](https://mlsanity.com/paper/2207.06983/) (2022)
- [AccoMontage: Accompaniment Arrangement via Phrase Selection and Style Transfer](https://mlsanity.com/paper/2108.11213/) (2021)
- [MELONS: generating melody with long-term structure using transformers and structure graph](https://mlsanity.com/paper/2110.05020/) (2021)
- [Compose & Embellish: Well-Structured Piano Performance Generation via A Two-Stage Approach](https://mlsanity.com/paper/2209.08212/) (2022)
- [Generating Lead Sheets with Affect: A Novel Conditional seq2seq Framework](https://mlsanity.com/paper/2104.13056/) (2021)
- [Narrative Text Generation with a Latent Discrete Plan](https://mlsanity.com/paper/2010.03272/) (2020)
- [BERT-like Pre-training for Symbolic Piano Music Classification Tasks](https://mlsanity.com/paper/2107.05223/) (2021)
- [Style Pooling: Automatic Text Style Obfuscation for Improved Classification Fairness](https://mlsanity.com/paper/2109.04624/) (2021)
- [A database linking piano and orchestral MIDI scores with application to automatic projective orchestration](https://mlsanity.com/paper/1810.08611/) (2018)
- [AccoMontage2: A Complete Harmonization and Accompaniment Arrangement System](https://mlsanity.com/paper/2209.00353/) (2022)
- [Rhythm, Chord and Melody Generation for Lead Sheets using Recurrent Neural Networks](https://mlsanity.com/paper/2002.10266/) (2020)

## Abstract (arXiv)

Lead sheets have become commonplace in generative music research, being used as an initial compressed representation for downstream tasks like multitrack music generation and automatic arrangement. Despite this, researchers have often fallen back on deterministic reduction methods (such as the skyline algorithm) to generate lead sheets when seeking paired lead sheets and full scores, with little attention being paid toward the quality of the lead sheets themselves and how they accurately reflect their orchestrated counterparts. To address these issues, we propose the problem of conditional lead sheet generation (i.e. generating a lead sheet given its full score version), and show that this task can be formulated as an unsupervised music compression task, where the lead sheet represents a compressed latent version of the score. We introduce a novel model, called Lead-AE, that models the lead sheets as a discrete subselection of the original sequence, using a differentiable top-k operator to allow for controllable local sparsity constraints. Across both automatic proxy tasks and direct human evaluations, we find that our method improves upon the established deterministic baseline and produces coherent reductions of large multitrack scores.