Structure-informed Positional Encoding for Music Generation - Archive ouverte HAL Access content directly
Conference Papers Year : 2024

Structure-informed Positional Encoding for Music Generation

Abstract

Music generated by deep learning methods often suffers from a lack of coherence and long-term organization. Yet, multi-scale hierarchical structure is a distinctive feature of music signals. To leverage this information, we propose a structure-informed positional encoding framework for music generation with Transformers. We design three variants in terms of absolute, relative and non-stationary positional information. We comprehensively test them on two symbolic music generation tasks: next-timestep prediction and accompaniment generation. As a comparison, we choose multiple baselines from the literature and demonstrate the merits of our methods using several musically-motivated evaluation metrics. In particular, our methods improve the melodic and structural consistency of the generated pieces.
Fichier principal
Vignette du fichier
svbwdvrdnrztpzxgdsckkhqxkjbjpfzx.pdf (1.16 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04432659 , version 1 (15-02-2024)
hal-04432659 , version 2 (20-02-2024)
hal-04432659 , version 3 (28-02-2024)

Identifiers

Cite

Manvi Agarwal, Changhong Wang, Gaël Richard. Structure-informed Positional Encoding for Music Generation. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2024, Seoul, South Korea. ⟨hal-04432659v3⟩
254 View
99 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More