Aligning Language Models for Lyric-to-Melody Generation with Rule-Based Musical Constraints

• URL: http://arxiv.org/abs/2604.18489v1
• Date: Mon, 20 Apr 2026 16:40:45 GMT
• Title: Aligning Language Models for Lyric-to-Melody Generation with Rule-Based Musical Constraints
• Abstract summary: Large Language Models (LLMs) show promise in lyric-to-melody generation, but models trained with Supervised Fine-Tuning (SFT) often produce implausible melodies.<n>We propose a novel alignment framework that instills musical knowledge without human annotation.
• Score: 9.07848158346737
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) show promise in lyric-to-melody generation, but models trained with Supervised Fine-Tuning (SFT) often produce musically implausible melodies with issues like poor rhythm and unsuitable vocal ranges, a phenomenon we term "constraint violation". To address this, we propose a novel alignment framework that instills musical knowledge without human annotation. We define rule-based musical constraints to automatically generate a preference dataset from an SFT model's outputs. The model is then aligned through a sequential process, first using Direct Preference Optimization (DPO) on paired preference data, followed by Kahneman-Tversky Optimization (KTO) on unpaired negative samples. Experimental results demonstrate that our aligned model substantially reduces rule violations and outperforms strong baselines in both objective and subjective evaluations, generating melodies with substantially improved musicality and coherence. An interactive demo with audio comparisons is available at https://arain233.github.io/AligningMelody-demo.

Related papers

• Towards Effective Negation Modeling in Joint Audio-Text Models for Music [3.7723788828505125]
  Joint audio-text models struggle with semantic phenomena such as negation.<n>We introduce negation through text augmentation and a dissimilarity-based contrastive loss.<n>We propose two protocols that frame negation modeling as retrieval and binary classification tasks.
  arXiv  Detail & Related papers  (2026-01-20T13:06:48Z)
• YingMusic-Singer: Zero-shot Singing Voice Synthesis and Editing with Annotation-free Melody Guidance [16.462715982402884]
  Singing Voice Synthesis (SVS) remains constrained in practical deployment due to its strong dependence on accurate phoneme-level alignment.<n>We propose a melody-driven SVS framework capable of synthesizing arbitrary lyrics following any reference melody.<n>Our method builds on a Diffusion Transformer (DiT) architecture, enhanced with a dedicated melody extraction module.
  arXiv  Detail & Related papers  (2025-12-04T13:25:33Z)
• JAM: A Tiny Flow-based Song Generator with Fine-grained Controllability and Aesthetic Alignment [26.590667516155083]
  Diffusion and flow-matching models have revolutionized automatic text-to-audio generation.<n>Recent open lyrics-to-song models have set an acceptable standard in automatic song generation for recreational use.<n>Our flow-matching-based JAM is the first effort toward endowing word-level timing and duration control in song generation.
  arXiv  Detail & Related papers  (2025-07-28T14:34:02Z)
• Scaling Self-Supervised Representation Learning for Symbolic Piano Performance [52.661197827466886]
  We study the capabilities of generative autoregressive transformer models trained on large amounts of symbolic solo-piano transcriptions.<n>We use a comparatively smaller, high-quality subset to finetune models to produce musical continuations, perform symbolic classification tasks, and produce general-purpose contrastive MIDI embeddings.
  arXiv  Detail & Related papers  (2025-06-30T14:00:14Z)
• LeVo: High-Quality Song Generation with Multi-Preference Alignment [47.965028296133426]
  We introduce LeVo, a language model based framework consisting of LeLM and Music Codec.<n>LeVo is capable of parallel modeling of two types of tokens: mixed tokens, which represent the combined audio of vocals and accompaniment.<n>It employs two decoder-only transformers and a modular extension training strategy to prevent interference between different token types.
  arXiv  Detail & Related papers  (2025-06-09T07:57:24Z)
• DITTO: Diffusion Inference-Time T-Optimization for Music Generation [49.90109850026932]
  Diffusion Inference-Time T-Optimization (DITTO) is a frame-work for controlling pre-trained text-to-music diffusion models at inference-time. We demonstrate a surprisingly wide-range of applications for music generation including inpainting, outpainting, and looping as well as intensity, melody, and musical structure control.
  arXiv  Detail & Related papers  (2024-01-22T18:10:10Z)
• RMSSinger: Realistic-Music-Score based Singing Voice Synthesis [56.51475521778443]
  RMS-SVS aims to generate high-quality singing voices given realistic music scores with different note types. We propose RMSSinger, the first RMS-SVS method, which takes realistic music scores as input. In RMSSinger, we introduce word-level modeling to avoid the time-consuming phoneme duration annotation and the complicated phoneme-level mel-note alignment.
  arXiv  Detail & Related papers  (2023-05-18T03:57:51Z)
• Unsupervised Melody-Guided Lyrics Generation [84.22469652275714]
  We propose to generate pleasantly listenable lyrics without training on melody-lyric aligned data. We leverage the crucial alignments between melody and lyrics and compile the given melody into constraints to guide the generation process.
  arXiv  Detail & Related papers  (2023-05-12T20:57:20Z)
• Re-creation of Creations: A New Paradigm for Lyric-to-Melody Generation [158.54649047794794]
  Re-creation of Creations (ROC) is a new paradigm for lyric-to-melody generation. ROC achieves good lyric-melody feature alignment in lyric-to-melody generation.
  arXiv  Detail & Related papers  (2022-08-11T08:44:47Z)
• Generating Music with a Self-Correcting Non-Chronological Autoregressive Model [6.289267097017553]
  We describe a novel approach for generating music using a self-correcting, non-chronological, autoregressive model. We represent music as a sequence of edit events, each of which denotes either the addition or removal of a note. During inference, we generate one edit event at a time using direct ancestral sampling.
  arXiv  Detail & Related papers  (2020-08-18T20:36:47Z)
• Continuous Melody Generation via Disentangled Short-Term Representations and Structural Conditions [14.786601824794369]
  We present a model for composing melodies given a user specified symbolic scenario combined with a previous music context. Our model is capable of generating long melodies by regarding 8-beat note sequences as basic units, and shares consistent rhythm pattern structure with another specific song. Results show that the music generated by our model tends to have salient repetition structures, rich motives, and stable rhythm patterns.
  arXiv  Detail & Related papers  (2020-02-05T06:23:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.