# An Comparative Analysis of Different Pitch and Metrical Grid Encoding Methods in the Task of Sequential Music Generation

arXiv: https://arxiv.org/abs/2301.13383 · HTML: https://mlsanity.com/paper/2301.13383/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 41 references that are in the map.
- First posted on arXiv 31 Jan 2023.

## Builds on (most cited first, 23 of 23)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://mlsanity.com/paper/1901.02860/) (2019)
- [Music Transformer](https://mlsanity.com/paper/1809.04281/) (2018)
- [MuseGAN: Multi-track Sequential Generative Adversarial Networks for Symbolic Music Generation and Accompaniment](https://mlsanity.com/paper/1709.06298/) (2017)
- [C-RNN-GAN: Continuous recurrent neural networks with adversarial training](https://mlsanity.com/paper/1611.09904/) (2016)
- [A Hierarchical Latent Vector Model for Learning Long-Term Structure in Music](https://mlsanity.com/paper/1803.05428/) (2018)
- [Generalization in Generation: A closer look at Exposure Bias](https://mlsanity.com/paper/1910.00292/) (2019)
- [Compound Word Transformer: Learning to Compose Full-Song Music over Dynamic Directed Hypergraphs](https://mlsanity.com/paper/2101.02402/) (2021)
- [DeepBach: a Steerable Model for Bach Chorales Generation](https://mlsanity.com/paper/1612.01010/) (2016)
- [MusicBERT: Symbolic Music Understanding with Large-Scale Pre-Training](https://mlsanity.com/paper/2106.05630/) (2021)
- [MidiNet: A Convolutional Generative Adversarial Network for Symbolic-domain Music Generation](https://mlsanity.com/paper/1703.10847/) (2017)
- [The Jazz Transformer on the Front Line: Exploring the Shortcomings of AI-composed Music through Quantitative Measures](https://mlsanity.com/paper/2008.01307/) (2020)
- [Deep Learning Techniques for Music Generation -- A Survey](https://mlsanity.com/paper/1709.01620/) (2017)
- [Scheduled Sampling for Transformers](https://mlsanity.com/paper/1906.07651/) (2019)
- [MusPy: A Toolkit for Symbolic Music Generation](https://mlsanity.com/paper/2008.01951/) (2020)
- [A Functional Taxonomy of Music Generation Systems](https://mlsanity.com/paper/1812.04186/) (2018)
- [Controllable deep melody generation via hierarchical music structure representation](https://mlsanity.com/paper/2109.00663/) (2021)
- [PIANOTREE VAE: Structured Representation Learning for Polyphonic Music](https://mlsanity.com/paper/2008.07118/) (2020)
- [PiRhDy: Learning Pitch-, Rhythm-, and Dynamics-aware Embeddings for Symbolic Music](https://mlsanity.com/paper/2010.08091/) (2020)
- [A Domain-Knowledge-Inspired Music Embedding Space and a Novel Attention Mechanism for Symbolic Music Modeling](https://mlsanity.com/paper/2212.00973/) (2022)
- [MeloForm: Generating Melody with Musical Form based on Expert Systems and Neural Networks](https://mlsanity.com/paper/2208.14345/) (2022)
- [SinTra: Learning an inspiration model from a single multi-track music segment](https://mlsanity.com/paper/2204.09917/) (2022)
- [Learning Transposition-Invariant Interval Features from Symbolic Music and Audio](https://mlsanity.com/paper/1806.08236/) (2018)

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

- [Natural Language Processing Methods for Symbolic Music Generation and Information Retrieval: a Survey](https://mlsanity.com/paper/2402.17467/) (2024)
- [Emotion-driven Piano Music Generation via Two-stage Disentanglement and Functional Representation](https://mlsanity.com/paper/2407.20955/) (2024)

## Abstract (arXiv)

Pitch and meter are two fundamental music features for symbolic music generation tasks, where researchers usually choose different encoding methods depending on specific goals. However, the advantages and drawbacks of different encoding methods have not been frequently discussed. This paper presents a integrated analysis of the influence of two low-level feature, pitch and meter, on the performance of a token-based sequential music generation model. First, the commonly used MIDI number encoding and a less used class-octave encoding are compared. Second, an dense intra-bar metric grid is imposed to the encoded sequence as auxiliary features. Different complexity and resolutions of the metric grid are compared. For complexity, the single token approach and the multiple token approach are compared; for grid resolution, 0 (ablation), 1 (bar-level), 4 (downbeat-level) 12, (8th-triplet-level) up to 64 (64th-note-grid-level) are compared; for duration resolution, 4, 8, 12 and 16 subdivisions per beat are compared. All different encodings are tested on separately trained Transformer-XL models for a melody generation task. Regarding distribution similarity of several objective evaluation metrics to the test dataset, results suggest that the class-octave encoding significantly outperforms the taken-for-granted MIDI encoding on pitch-related metrics; finer grids and multiple-token grids improve the rhythmic quality, but also suffer from over-fitting at early training stage. Results display a general phenomenon of over-fitting from two aspects, the pitch embedding space and the test loss of the single-token grid encoding. From a practical perspective, we both demonstrate the feasibility and raise the concern of easy over-fitting problem of using smaller networks and lower embedding dimensions on the generation task. The findings can also contribute to futural models in terms of feature engineering.