Unified Cross-modal Translation of Score Images, Symbolic Music, and Performance Audio

• URL: http://arxiv.org/abs/2505.12863v1
• Date: Mon, 19 May 2025 08:46:45 GMT
• Title: Unified Cross-modal Translation of Score Images, Symbolic Music, and Performance Audio
• Authors: Jongmin Jung, Dongmin Kim, Sihun Lee, Seola Cho, Hyungjoon Soh, Irmak Bukey, Chris Donahue, Dasaem Jeong,
• Abstract summary: We train a general-purpose model on many translation tasks simultaneously.<n>We propose a new large-scale dataset and the tokenization of each modality.<n>Our approach achieves the first successful score-image-conditioned audio generation.
• Score: 7.518711227993383
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Music exists in various modalities, such as score images, symbolic scores, MIDI, and audio. Translations between each modality are established as core tasks of music information retrieval, such as automatic music transcription (audio-to-MIDI) and optical music recognition (score image to symbolic score). However, most past work on multimodal translation trains specialized models on individual translation tasks. In this paper, we propose a unified approach, where we train a general-purpose model on many translation tasks simultaneously. Two key factors make this unified approach viable: a new large-scale dataset and the tokenization of each modality. Firstly, we propose a new dataset that consists of more than 1,300 hours of paired audio-score image data collected from YouTube videos, which is an order of magnitude larger than any existing music modal translation datasets. Secondly, our unified tokenization framework discretizes score images, audio, MIDI, and MusicXML into a sequence of tokens, enabling a single encoder-decoder Transformer to tackle multiple cross-modal translation as one coherent sequence-to-sequence task. Experimental results confirm that our unified multitask model improves upon single-task baselines in several key areas, notably reducing the symbol error rate for optical music recognition from 24.58% to a state-of-the-art 13.67%, while similarly substantial improvements are observed across the other translation tasks. Notably, our approach achieves the first successful score-image-conditioned audio generation, marking a significant breakthrough in cross-modal music generation.

Related papers

• AudioX: Diffusion Transformer for Anything-to-Audio Generation [72.84633243365093]
  AudioX is a unified Diffusion Transformer model for Anything-to-Audio and Music Generation.<n>It can generate both general audio and music with high quality, while offering flexible natural language control.<n>To address data scarcity, we curate two datasets: vggsound-caps with 190K audio captions based on the VGGSound dataset, and V2M-caps with 6 million music captions derived from the V2M dataset.
  arXiv  Detail & Related papers  (2025-03-13T16:30:59Z)
• UniMuMo: Unified Text, Music and Motion Generation [57.72514622935806]
  We introduce UniMuMo, a unified multimodal model capable of taking arbitrary text, music, and motion data as input conditions to generate outputs across all three modalities. By converting music, motion, and text into token-based representation, our model bridges these modalities through a unified encoder-decoder transformer architecture.
  arXiv  Detail & Related papers  (2024-10-06T16:04:05Z)
• PerTok: Expressive Encoding and Modeling of Symbolic Musical Ideas and Variations [0.3683202928838613]
  Cadenza is a new multi-stage generative framework for predicting expressive variations of symbolic musical ideas. The proposed framework comprises of two sequential stages: 1) Composer and 2) Performer. Our framework is designed, researched and implemented with the objective of providing inspiration for musicians.
  arXiv  Detail & Related papers  (2024-10-02T22:11:31Z)
• MMTrail: A Multimodal Trailer Video Dataset with Language and Music Descriptions [69.9122231800796]
  We present MMTrail, a large-scale multi-modality video-language dataset incorporating more than 20M trailer clips with visual captions.<n>We propose a systemic captioning framework, achieving various modality annotations with more than 27.1k hours of trailer videos.<n>Our dataset potentially paves the path for fine-grained large multimodal-language model training.
  arXiv  Detail & Related papers  (2024-07-30T16:43:24Z)
• Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models [98.34889301515412]
  We develop the Qwen-Audio model and address the limitation by scaling up audio-language pre-training to cover over 30 tasks and various audio types. Qwen-Audio achieves impressive performance across diverse benchmark tasks without requiring any task-specific fine-tuning. We further develop Qwen-Audio-Chat, which allows for input from various audios and text inputs, enabling multi-turn dialogues and supporting various audio-central scenarios.
  arXiv  Detail & Related papers  (2023-11-14T05:34:50Z)
• Timbre-Trap: A Low-Resource Framework for Instrument-Agnostic Music Transcription [19.228155694144995]
  Timbre-Trap is a novel framework which unifies music transcription and audio reconstruction. We train a single autoencoder to simultaneously estimate pitch salience and reconstruct complex spectral coefficients. We demonstrate that the framework leads to performance comparable to state-of-the-art instrument-agnostic transcription methods.
  arXiv  Detail & Related papers  (2023-09-27T15:19:05Z)
• Simple and Controllable Music Generation [94.61958781346176]
  MusicGen is a single Language Model (LM) that operates over several streams of compressed discrete music representation, i.e., tokens. Unlike prior work, MusicGen is comprised of a single-stage transformer LM together with efficient token interleaving patterns.
  arXiv  Detail & Related papers  (2023-06-08T15:31:05Z)
• An Comparative Analysis of Different Pitch and Metrical Grid Encoding Methods in the Task of Sequential Music Generation [4.941630596191806]
  This paper presents an analysis of the influence of pitch and meter on the performance of a token-based sequential music generation model. 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) up to 16 subdivisions per beat are compared. Results suggest that the class-octave encoding significantly outperforms the taken-for-granted MIDI encoding on pitch-related metrics.
  arXiv  Detail & Related papers  (2023-01-31T03:19:50Z)
• Museformer: Transformer with Fine- and Coarse-Grained Attention for Music Generation [138.74751744348274]
  We propose Museformer, a Transformer with a novel fine- and coarse-grained attention for music generation. Specifically, with the fine-grained attention, a token of a specific bar directly attends to all the tokens of the bars that are most relevant to music structures. With the coarse-grained attention, a token only attends to the summarization of the other bars rather than each token of them so as to reduce the computational cost.
  arXiv  Detail & Related papers  (2022-10-19T07:31:56Z)

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.