pith. sign in

arxiv: 1809.04281 · v3 · pith:R6YUL47Ynew · submitted 2018-09-12 · 💻 cs.LG · cs.SD· eess.AS· stat.ML

Music Transformer

classification 💻 cs.LG cs.SDeess.ASstat.ML
keywords relativetransformermusicattentiongeneratelengthsequencestructure
0
0 comments X
read the original abstract

Music relies heavily on repetition to build structure and meaning. Self-reference occurs on multiple timescales, from motifs to phrases to reusing of entire sections of music, such as in pieces with ABA structure. The Transformer (Vaswani et al., 2017), a sequence model based on self-attention, has achieved compelling results in many generation tasks that require maintaining long-range coherence. This suggests that self-attention might also be well-suited to modeling music. In musical composition and performance, however, relative timing is critically important. Existing approaches for representing relative positional information in the Transformer modulate attention based on pairwise distance (Shaw et al., 2018). This is impractical for long sequences such as musical compositions since their memory complexity for intermediate relative information is quadratic in the sequence length. We propose an algorithm that reduces their intermediate memory requirement to linear in the sequence length. This enables us to demonstrate that a Transformer with our modified relative attention mechanism can generate minute-long compositions (thousands of steps, four times the length modeled in Oore et al., 2018) with compelling structure, generate continuations that coherently elaborate on a given motif, and in a seq2seq setup generate accompaniments conditioned on melodies. We evaluate the Transformer with our relative attention mechanism on two datasets, JSB Chorales and Piano-e-Competition, and obtain state-of-the-art results on the latter.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. ONOTE: Benchmarking Omnimodal Notation Processing for Expert-level Music Intelligence

    cs.SD 2026-04 unverdicted novelty 7.0

    ONOTE is a multi-format benchmark that applies a deterministic pipeline to expose a disconnect between perceptual accuracy and music-theoretic comprehension in leading omnimodal AI models.

  2. Generating Long Sequences with Sparse Transformers

    cs.LG 2019-04 unverdicted novelty 7.0

    Sparse Transformers factorize attention to handle sequences tens of thousands long, achieving new SOTA density modeling on Enwik8, CIFAR-10, and ImageNet-64.

  3. PJ-RoPE: A Fourier-Jet-Affine Position Space for Relative Attention

    cs.LG 2026-06 unverdicted novelty 6.0

    PJ-RoPE organizes relative-position mechanisms as a learnable Fourier-Jet-Affine space derived from lag-shift dynamics, extending RoPE and ALiBi with explicit jets and sector selection.

  4. Scaling Data-Constrained Language Models

    cs.CL 2023-05 conditional novelty 6.0

    Repeating training data up to 4 epochs yields negligible loss increase versus unique data for fixed compute, and a new scaling law accounts for the decaying value of repeated tokens and excess parameters.

  5. A Bi-directional Transformer for Musical Chord Recognition

    cs.SD 2019-07 unverdicted novelty 6.0

    A bi-directional Transformer achieves competitive chord recognition by using self-attention to capture long-term dependencies in audio in a single training phase.

  6. From Image to Music Language: A Two-Stage Structure Decoding Approach for Complex Polyphonic OMR

    cs.SD 2026-04 unverdicted novelty 5.0

    A two-stage OMR pipeline decodes symbol candidates into polyphonic score structures via topology recognition with probability-guided search.

  7. Mathematical Foundations of Polyphonic Music Generation via Structural Inductive Bias

    cs.LG 2026-01 unverdicted novelty 5.0

    Smart Embedding reduces parameters by 48.3 percent in polyphonic music models with information-theoretic loss bounds under 0.153 bits and tighter generalization via Rademacher complexity.

  8. In Context Learning and Reasoning for Symbolic Regression with Large Language Models

    cs.CL 2024-10 unverdicted novelty 5.0

    GPT-4 models rediscover Langmuir isotherms and produce fits on Nikuradse pipe-flow data via iterative chain-of-thought prompting with scientific context and external code feedback.

  9. Agglomerative Attention

    cs.LG 2019-07 unverdicted novelty 5.0

    Presents agglomerative attention, a linear-complexity attention model that achieves comparable performance to full attention on language modeling tasks.

  10. Real-Time Language Model Jamming: A Case Study for Live Music Accompaniment Generation

    cs.SD 2026-06 unverdicted novelty 4.0

    StreamMUSE performs frame-synchronous streaming inference for language models by having a client send high-frequency requests and a server return outputs aligned to an external clock, shown on live music accompaniment...

  11. Musical Attention Transformer: Music Generation Using a Music-Specific Attention Model

    cs.SD 2026-05 unverdicted novelty 4.0

    The paper introduces Musical Attention, an attention variant that incorporates eight musical features including metadata to generate more coherent and varied music than standard or strided attention baselines.