Gus Xia

Ruibin Yuan, Yinghao Ma, Yizhi Li et al. · deepmind, mila

In the era of extensive intersection between art and Artificial Intelligence (AI), such as image generation and fiction co-creation, AI for music remains relatively nascent, particularly in music understanding. This is evident in the limited work on deep music representations, the scarcity of large-scale datasets, and the absence of a universal and community-driven benchmark. To address this issue, we introduce the Music Audio Representation Benchmark for universaL Evaluation, termed MARBLE. It aims to provide a benchmark for various Music Information Retrieval (MIR) tasks by defining a comprehensive taxonomy with four hierarchy levels, including acoustic, performance, score, and high-level description. We then establish a unified protocol based on 14 tasks on 8 public-available datasets, providing a fair and standard assessment of representations of all open-sourced pre-trained models developed on music recordings as baselines. Besides, MARBLE offers an easy-to-use, extendable, and reproducible suite for the community, with a clear statement on copyright issues on datasets. Results suggest recently proposed large-scale pre-trained musical language models perform the best in most tasks, with room for further improvement. The leaderboard and toolkit repository are published at https://marble-bm.shef.ac.uk to promote future music AI research.

Yinghao Ma, Ruibin Yuan, Yizhi Li et al. · deepmind, mila

Self-supervised learning (SSL) has shown promising results in various speech and natural language processing applications. However, its efficacy in music information retrieval (MIR) still remains largely unexplored. While previous SSL models pre-trained on music recordings may have been mostly closed-sourced, recent speech models such as wav2vec2.0 have shown promise in music modelling. Nevertheless, research exploring the effectiveness of applying speech SSL models to music recordings has been limited. We explore the music adaption of SSL with two distinctive speech-related models, data2vec1.0 and Hubert, and refer to them as music2vec and musicHuBERT, respectively. We train $12$ SSL models with 95M parameters under various pre-training configurations and systematically evaluate the MIR task performances with 13 different MIR tasks. Our findings suggest that training with music data can generally improve performance on MIR tasks, even when models are trained using paradigms designed for speech. However, we identify the limitations of such existing speech-oriented designs, especially in modelling polyphonic information. Based on the experimental results, empirical suggestions are also given for designing future musical SSL strategies and paradigms.

Yixiao Zhang, Junyan Jiang, Gus Xia et al. · bytedance

Lyric interpretations can help people understand songs and their lyrics quickly, and can also make it easier to manage, retrieve and discover songs efficiently from the growing mass of music archives. In this paper we propose BART-fusion, a novel model for generating lyric interpretations from lyrics and music audio that combines a large-scale pre-trained language model with an audio encoder. We employ a cross-modal attention module to incorporate the audio representation into the lyrics representation to help the pre-trained language model understand the song from an audio perspective, while preserving the language model's original generative performance. We also release the Song Interpretation Dataset, a new large-scale dataset for training and evaluating our model. Experimental results show that the additional audio information helps our model to understand words and music better, and to generate precise and fluent interpretations. An additional experiment on cross-modal music retrieval shows that interpretations generated by BART-fusion can also help people retrieve music more accurately than with the original BART.

Liwei Lin, Gus Xia, Junyan Jiang et al. · bytedance

Recent years have witnessed a rapid growth of large-scale language models in the domain of music audio. Such models enable end-to-end generation of higher-quality music, and some allow conditioned generation using text descriptions. However, the control power of text controls on music is intrinsically limited, as they can only describe music indirectly through meta-data (such as singers and instruments) or high-level representations (such as genre and emotion). We aim to further equip the models with direct and content-based controls on innate music languages such as pitch, chords and drum track. To this end, we contribute Coco-Mulla, a content-based control method for music large language modeling. It uses a parameter-efficient fine-tuning (PEFT) method tailored for Transformer-based audio models. Experiments show that our approach achieved high-quality music generation with low-resource semi-supervised learning, tuning with less than 4% parameters compared to the original model and training on a small dataset with fewer than 300 songs. Moreover, our approach enables effective content-based controls, and we illustrate the control power via chords and rhythms, two of the most salient features of music audio. Furthermore, we show that by combining content-based controls and text descriptions, our system achieves flexible music variation generation and arrangement. Our source codes and demos are available online.

Yixiao Zhang, Akira Maezawa, Gus Xia et al. · bytedance

Creating music is iterative, requiring varied methods at each stage. However, existing AI music systems fall short in orchestrating multiple subsystems for diverse needs. To address this gap, we introduce Loop Copilot, a novel system that enables users to generate and iteratively refine music through an interactive, multi-round dialogue interface. The system uses a large language model to interpret user intentions and select appropriate AI models for task execution. Each backend model is specialized for a specific task, and their outputs are aggregated to meet the user's requirements. To ensure musical coherence, essential attributes are maintained in a centralized table. We evaluate the effectiveness of the proposed system through semi-structured interviews and questionnaires, highlighting its utility not only in facilitating music creation but also its potential for broader applications.

Yinghao Ma, Anders Øland, Anton Ragni et al.

In recent years, foundation models (FMs) such as large language models (LLMs) and latent diffusion models (LDMs) have profoundly impacted diverse sectors, including music. This comprehensive review examines state-of-the-art (SOTA) pre-trained models and foundation models in music, spanning from representation learning, generative learning and multimodal learning. We first contextualise the significance of music in various industries and trace the evolution of AI in music. By delineating the modalities targeted by foundation models, we discover many of the music representations are underexplored in FM development. Then, emphasis is placed on the lack of versatility of previous methods on diverse music applications, along with the potential of FMs in music understanding, generation and medical application. By comprehensively exploring the details of the model pre-training paradigm, architectural choices, tokenisation, finetuning methodologies and controllability, we emphasise the important topics that should have been well explored, like instruction tuning and in-context learning, scaling law and emergent ability, as well as long-sequence modelling etc. A dedicated section presents insights into music agents, accompanied by a thorough analysis of datasets and evaluations essential for pre-training and downstream tasks. Finally, by underscoring the vital importance of ethical considerations, we advocate that following research on FM for music should focus more on such issues as interpretability, transparency, human responsibility, and copyright issues. The paper offers insights into future challenges and trends on FMs for music, aiming to shape the trajectory of human-AI collaboration in the music realm.

Runbang Zhang, Yixiao Zhang, Kai Shao et al. · bytedance

In this study, we explore the representation mapping from the domain of visual arts to the domain of music, with which we can use visual arts as an effective handle to control music generation. Unlike most studies in multimodal representation learning that are purely data-driven, we adopt an analysis-by-synthesis approach that combines deep music representation learning with user studies. Such an approach enables us to discover \textit{interpretable} representation mapping without a huge amount of paired data. In particular, we discover that visual-to-music mapping has a nice property similar to equivariant. In other words, we can use various image transformations, say, changing brightness, changing contrast, style transfer, to control the corresponding transformations in the music domain. In addition, we released the Vis2Mus system as a controllable interface for symbolic music generation.

Junyan Jiang, Daniel Chin, Yixiao Zhang et al. · bytedance

Music contains hierarchical structures beyond beats and measures. While hierarchical structure annotations are helpful for music information retrieval and computer musicology, such annotations are scarce in current digital music databases. In this paper, we explore a data-driven approach to automatically extract hierarchical metrical structures from scores. We propose a new model with a Temporal Convolutional Network-Conditional Random Field (TCN-CRF) architecture. Given a symbolic music score, our model takes in an arbitrary number of voices in a beat-quantized form, and predicts a 4-level hierarchical metrical structure from downbeat-level to section-level. We also annotate a dataset using RWC-POP MIDI files to facilitate training and evaluation. We show by experiments that the proposed method performs better than the rule-based approach under different orchestration settings. We also perform some simple musicological analysis on the model predictions. All demos, datasets and pre-trained models are publicly available on Github.

Lejun Min, Junyan Jiang, Gus Xia et al.

We propose Polyffusion, a diffusion model that generates polyphonic music scores by regarding music as image-like piano roll representations. The model is capable of controllable music generation with two paradigms: internal control and external control. Internal control refers to the process in which users pre-define a part of the music and then let the model infill the rest, similar to the task of masked music generation (or music inpainting). External control conditions the model with external yet related information, such as chord, texture, or other features, via the cross-attention mechanism. We show that by using internal and external controls, Polyffusion unifies a wide range of music creation tasks, including melody generation given accompaniment, accompaniment generation given melody, arbitrary music segment inpainting, and music arrangement given chords or textures. Experimental results show that our model significantly outperforms existing Transformer and sampling-based baselines, and using pre-trained disentangled representations as external conditions yields more effective controls.

Jingwei Zhao, Gus Xia, Ziyu Wang et al.

In the realm of music AI, arranging rich and structured multi-track accompaniments from a simple lead sheet presents significant challenges. Such challenges include maintaining track cohesion, ensuring long-term coherence, and optimizing computational efficiency. In this paper, we introduce a novel system that leverages prior modelling over disentangled style factors to address these challenges. Our method presents a two-stage process: initially, a piano arrangement is derived from the lead sheet by retrieving piano texture styles; subsequently, a multi-track orchestration is generated by infusing orchestral function styles into the piano arrangement. Our key design is the use of vector quantization and a unique multi-stream Transformer to model the long-term flow of the orchestration style, which enables flexible, controllable, and structured music generation. Experiments show that by factorizing the arrangement task into interpretable sub-stages, our approach enhances generative capacity while improving efficiency. Additionally, our system supports a variety of music genres and provides style control at different composition hierarchies. We further show that our system achieves superior coherence, structure, and overall arrangement quality compared to existing baselines.

Yuxuan Wu, Roger B. Dannenberg, Gus Xia

Music motif, as a conceptual building block of composition, is crucial for music structure analysis and automatic composition. While human listeners can identify motifs easily, existing computational models fall short in representing motifs and their developments. The reason is that the nature of motifs is implicit, and the diversity of motif variations extends beyond simple repetitions and modulations. In this study, we aim to learn the implicit relationship between motifs and their variations via representation learning, using the Siamese network architecture and a pretraining and fine-tuning pipeline. A regularization-based method, VICReg, is adopted for pretraining, while contrastive learning is used for fine-tuning. Experimental results on a retrieval-based task show that these two methods complement each other, yielding an improvement of 12.6% in the area under the precision-recall curve. Lastly, we visualize the acquired motif representations, offering an intuitive comprehension of the overall structure of a music piece. As far as we know, this work marks a noteworthy step forward in computational modeling of music motifs. We believe that this work lays the foundations for future applications of motifs in automatic music composition and music information retrieval.

Xuanjie Liu, Daniel Chin, Yichen Huang et al.

We have recently seen great progress in learning interpretable music representations, ranging from basic factors, such as pitch and timbre, to high-level concepts, such as chord and texture. However, most methods rely heavily on music domain knowledge. It remains an open question what general computational principles give rise to interpretable representations, especially low-dim factors that agree with human perception. In this study, we take inspiration from modern physics and use physical symmetry as a self consistency constraint for the latent space of time-series data. Specifically, it requires the prior model that characterises the dynamics of the latent states to be equivariant with respect to certain group transformations. We show that physical symmetry leads the model to learn a linear pitch factor from unlabelled monophonic music audio in a self-supervised fashion. In addition, the same methodology can be applied to computer vision, learning a 3D Cartesian space from videos of a simple moving object without labels. Furthermore, physical symmetry naturally leads to counterfactual representation augmentation, a new technique which improves sample efficiency.

Longshen Ou, Jingwei Zhao, Ziyu Wang et al.

We present a unified framework for automatic multitrack music arrangement that enables a single pre-trained symbolic music model to handle diverse arrangement scenarios, including reinterpretation, simplification, and additive generation. At its core is a segment-level reconstruction objective operating on token-level disentangled content and style, allowing for flexible any-to-any instrumentation transformations at inference time. To support track-wise modeling, we introduce REMI-z, a structured tokenization scheme for multitrack symbolic music that enhances modeling efficiency and effectiveness for both arrangement tasks and unconditional generation. Our method outperforms task-specific state-of-the-art models on representative tasks in different arrangement scenarios -- band arrangement, piano reduction, and drum arrangement, in both objective metrics and perceptual evaluations. Taken together, our framework demonstrates strong generality and suggests broader applicability in symbolic music-to-music transformation.

Yang Qu, Yutian Qin, Lecheng Chao et al.

The relationship between perceptual loudness and physical attributes of sound is an important subject in both computer music and psychoacoustics. Early studies of "equal-loudness contour" can trace back to the 1920s and the measured loudness with respect to intensity and frequency has been revised many times since then. However, most studies merely focus on synthesized sound, and the induced theories on natural tones with complex timbre have rarely been justified. To this end, we investigate both theory and applications of natural-tone loudness perception in this paper via modeling piano tone. The theory part contains: 1) an accurate measurement of piano-tone equal-loudness contour of pitches, and 2) a machine-learning model capable of inferring loudness purely based on spectral features trained on human subject measurements. As for the application, we apply our theory to piano control transfer, in which we adjust the MIDI velocities on two different player pianos (in different acoustic environments) to achieve the same perceptual effect. Experiments show that both our theoretical loudness modeling and the corresponding performance control transfer algorithm significantly outperform their baselines.

Ruibin Yuan, Hanfeng Lin, Yi Wang et al.

While Large Language Models (LLMs) demonstrate impressive capabilities in text generation, we find that their ability has yet to be generalized to music, humanity's creative language. We introduce ChatMusician, an open-source LLM that integrates intrinsic musical abilities. It is based on continual pre-training and finetuning LLaMA2 on a text-compatible music representation, ABC notation, and the music is treated as a second language. ChatMusician can understand and generate music with a pure text tokenizer without any external multi-modal neural structures or tokenizers. Interestingly, endowing musical abilities does not harm language abilities, even achieving a slightly higher MMLU score. Our model is capable of composing well-structured, full-length music, conditioned on texts, chords, melodies, motifs, musical forms, etc, surpassing GPT-4 baseline. On our meticulously curated college-level music understanding benchmark, MusicTheoryBench, ChatMusician surpasses LLaMA2 and GPT-3.5 on zero-shot setting by a noticeable margin. Our work reveals that LLMs can be an excellent compressor for music, but there remains significant territory to be conquered. We release our 4B token music-language corpora MusicPile, the collected MusicTheoryBench, code, model and demo in GitHub.

Yuxuan Wu, Ziyu Wang, Bhiksha Raj et al.

We contribute an unsupervised method that effectively learns disentangled content and style representations from sequences of observations. Unlike most disentanglement algorithms that rely on domain-specific labels or knowledge, our method is based on the insight of domain-general statistical differences between content and style -- content varies more among different fragments within a sample but maintains an invariant vocabulary across data samples, whereas style remains relatively invariant within a sample but exhibits more significant variation across different samples. We integrate such inductive bias into an encoder-decoder architecture and name our method after V3 (variance-versus-invariance). Experimental results show that V3 generalizes across multiple domains and modalities, successfully learning disentangled content and style representations, such as pitch and timbre from music audio, digit and color from images of hand-written digits, and action and character appearance from simple animations. V3 demonstrates strong disentanglement performance compared to existing unsupervised methods, along with superior out-of-distribution generalization under few-shot adaptation compared to supervised counterparts. Lastly, symbolic-level interpretability emerges in the learned content codebook, forging a near one-to-one alignment between machine representation and human knowledge.

Xingwei Qu, Yuelin Bai, Yinghao Ma et al.

In this paper, we explore the application of Large Language Models (LLMs) to the pre-training of music. While the prevalent use of MIDI in music modeling is well-established, our findings suggest that LLMs are inherently more compatible with ABC Notation, which aligns more closely with their design and strengths, thereby enhancing the model's performance in musical composition. To address the challenges associated with misaligned measures from different tracks during generation, we propose the development of a Synchronized Multi-Track ABC Notation (SMT-ABC Notation), which aims to preserve coherence across multiple musical tracks. Our contributions include a series of models capable of handling up to 8192 tokens, covering 90% of the symbolic music data in our training set. Furthermore, we explore the implications of the Symbolic Music Scaling Law (SMS Law) on model performance. The results indicate a promising direction for future research in music generation, offering extensive resources for community-led research through our open-source contributions.

Yixiao Zhang, Yukara Ikemiya, Gus Xia et al. · bytedance

Recent advances in text-to-music generation models have opened new avenues in musical creativity. However, music generation usually involves iterative refinements, and how to edit the generated music remains a significant challenge. This paper introduces a novel approach to the editing of music generated by such models, enabling the modification of specific attributes, such as genre, mood and instrument, while maintaining other aspects unchanged. Our method transforms text editing to \textit{latent space manipulation} while adding an extra constraint to enforce consistency. It seamlessly integrates with existing pretrained text-to-music diffusion models without requiring additional training. Experimental results demonstrate superior performance over both zero-shot and certain supervised baselines in style and timbre transfer evaluations. Additionally, we showcase the practical applicability of our approach in real-world music editing scenarios.

Ruibin Yuan, Hanfeng Lin, Shuyue Guo et al.

We tackle the task of long-form music generation--particularly the challenging \textbf{lyrics-to-song} problem--by introducing YuE, a family of open foundation models based on the LLaMA2 architecture. Specifically, YuE scales to trillions of tokens and generates up to five minutes of music while maintaining lyrical alignment, coherent musical structure, and engaging vocal melodies with appropriate accompaniment. It achieves this through (1) track-decoupled next-token prediction to overcome dense mixture signals, (2) structural progressive conditioning for long-context lyrical alignment, and (3) a multitask, multiphase pre-training recipe to converge and generalize. In addition, we redesign the in-context learning technique for music generation, enabling versatile style transfer (e.g., converting Japanese city pop into an English rap while preserving the original accompaniment) and bidirectional generation. Through extensive evaluation, we demonstrate that YuE matches or even surpasses some of the proprietary systems in musicality and vocal agility. In addition, fine-tuning YuE enables additional controls and enhanced support for tail languages. Furthermore, beyond generation, we show that YuE's learned representations can perform well on music understanding tasks, where the results of YuE match or exceed state-of-the-art methods on the MARBLE benchmark. Keywords: lyrics2song, song generation, long-form, foundation model, music generation

Huan Zhang, Taegyun Kwon, Anders Friburg et al.

This paper presents a comprehensive documentation of RenCon 2025, the revival of the expressive performance rendering competition which took place at ISMIR 2025 in Daejeon, Korea. The competition attracted 9 entries from international research groups, representing diverse approaches to expressive piano performance rendering. The two-phase assessment structure comprised a preliminary online evaluation and live real-time rendering at the conference. We analyze the competition format, participant demographics, system performance, and lessons learned for future iterations. The results demonstrate significant advances in expressive rendering capabilities while highlighting remaining challenges in achieving human-level musical expression.

Liwei Lin, Gus Xia, Yixiao Zhang et al. · bytedance

Controllable music generation plays a vital role in human-AI music co-creation. While Large Language Models (LLMs) have shown promise in generating high-quality music, their focus on autoregressive generation limits their utility in music editing tasks. To address this gap, we propose a novel approach leveraging a parameter-efficient heterogeneous adapter combined with a masking training scheme. This approach enables autoregressive language models to seamlessly address music inpainting tasks. Additionally, our method integrates frame-level content-based controls, facilitating track-conditioned music refinement and score-conditioned music arrangement. We apply this method to fine-tune MusicGen, a leading autoregressive music generation model. Our experiments demonstrate promising results across multiple music editing tasks, offering more flexible controls for future AI-driven music editing tools. The source codes and a demo page showcasing our work are available at https://kikyo-16.github.io/AIR.

Ziyu Wang, Lejun Min, Gus Xia

Recent deep music generation studies have put much emphasis on long-term generation with structures. However, we are yet to see high-quality, well-structured whole-song generation. In this paper, we make the first attempt to model a full music piece under the realization of compositional hierarchy. With a focus on symbolic representations of pop songs, we define a hierarchical language, in which each level of hierarchy focuses on the semantics and context dependency at a certain music scope. The high-level languages reveal whole-song form, phrase, and cadence, whereas the low-level languages focus on notes, chords, and their local patterns. A cascaded diffusion model is trained to model the hierarchical language, where each level is conditioned on its upper levels. Experiments and analysis show that our model is capable of generating full-piece music with recognizable global verse-chorus structure and cadences, and the music quality is higher than the baselines. Additionally, we show that the proposed model is controllable in a flexible way. By sampling from the interpretable hierarchical languages or adjusting pre-trained external representations, users can control the music flow via various features such as phrase harmonic structures, rhythmic patterns, and accompaniment texture.

Atharva Mehta, Shivam Chauhan, Amirbek Djanibekov et al.

The advent of Music-Language Models has greatly enhanced the automatic music generation capability of AI systems, but they are also limited in their coverage of the musical genres and cultures of the world. We present a study of the datasets and research papers for music generation and quantify the bias and under-representation of genres. We find that only 5.7% of the total hours of existing music datasets come from non-Western genres, which naturally leads to disparate performance of the models across genres. We then investigate the efficacy of Parameter-Efficient Fine-Tuning (PEFT) techniques in mitigating this bias. Our experiments with two popular models -- MusicGen and Mustango, for two underrepresented non-Western music traditions -- Hindustani Classical and Turkish Makam music, highlight the promises as well as the non-triviality of cross-genre adaptation of music through small datasets, implying the need for more equitable baseline music-language models that are designed for cross-cultural transfer learning.

Daniel Chin, Yuxuan Wang, Gus Xia

Large Language Model (LLM) -in-the-loop applications have been shown to effectively interpret the human user's commands, make plans, and operate external tools/systems accordingly. Still, the operation scope of the LLM agent is limited to passively following the user, requiring the user to frame his/her needs with regard to the underlying tools/systems. We note that the potential of an LLM-Agent User Interface (LAUI) is much greater. A user mostly ignorant to the underlying tools/systems should be able to work with a LAUI to discover an emergent workflow. Contrary to the conventional way of designing an explorable GUI to teach the user a predefined set of ways to use the system, in the ideal LAUI, the LLM agent is initialized to be proficient with the system, proactively studies the user and his/her needs, and proposes new interaction schemes to the user. To illustrate LAUI, we present Flute X GPT, a concrete example using an LLM agent, a prompt manager, and a flute-tutoring multi-modal software-hardware system to facilitate the complex, real-time user experience of learning to play the flute.

Megha Sharma, Muhammad Taimoor Haseeb, Gus Xia et al.

This paper introduces M2M Gen, a multi modal framework for generating background music tailored to Japanese manga. The key challenges in this task are the lack of an available dataset or a baseline. To address these challenges, we propose an automated music generation pipeline that produces background music for an input manga book. Initially, we use the dialogues in a manga to detect scene boundaries and perform emotion classification using the characters faces within a scene. Then, we use GPT4o to translate this low level scene information into a high level music directive. Conditioned on the scene information and the music directive, another instance of GPT 4o generates page level music captions to guide a text to music model. This produces music that is aligned with the mangas evolving narrative. The effectiveness of M2M Gen is confirmed through extensive subjective evaluations, showcasing its capability to generate higher quality, more relevant and consistent music that complements specific scenes when compared to our baselines.

Chenxi Wang, Yixuan Zhang, Ruiji Yu et al.

As the demand for emotional intelligence in large language models (LLMs) grows, a key challenge lies in understanding the internal mechanisms that give rise to emotional expression and in controlling emotions in generated text. This study addresses three core questions: (1) Do LLMs contain context-agnostic mechanisms shaping emotional expression? (2) What form do these mechanisms take? (3) Can they be harnessed for universal emotion control? We first construct a controlled dataset, SEV (Scenario-Event with Valence), to elicit comparable internal states across emotions. Subsequently, we extract context-agnostic emotion directions that reveal consistent, cross-context encoding of emotion (Q1). We identify neurons and attention heads that locally implement emotional computation through analytical decomposition and causal analysis, and validate their causal roles via ablation and enhancement interventions. Next, we quantify each sublayer's causal influence on the model's final emotion representation and integrate the identified local components into coherent global emotion circuits that drive emotional expression (Q2). Directly modulating these circuits achieves 99.65% emotion-expression accuracy on the test set, surpassing prompting- and steering-based methods (Q3). To our knowledge, this is the first systematic study to uncover and validate emotion circuits in LLMs, offering new insights into interpretability and controllable emotional intelligence.

Fathinah Izzati, Xinyue Li, Yuxuan Wu et al.

Humans can imagine various atmospheres and settings when listening to music, envisioning movie scenes that complement each piece. For example, slow, melancholic music might evoke scenes of heartbreak, while upbeat melodies suggest celebration. This paper explores whether a Music Language Model, e.g. MU-LLaMA, can perform a similar task, called Music Scene Imagination (MSI), which requires cross-modal information from video and music to train. To improve upon existing music captioning models which focusing solely on musical elements, we introduce MusiScene, a music captioning model designed to imagine scenes that complement each music. In this paper, (1) we construct a large-scale video-audio caption dataset with 3,371 pairs, (2) we finetune Music Understanding LLaMA for the MSI task to create MusiScene, and (3) we conduct comprehensive evaluations and prove that our MusiScene is more capable of generating contextually relevant captions compared to MU-LLaMA. We leverage the generated MSI captions to enhance Video Background Music Generation (VBMG) from text.

Fathinah Izzati, Xinyue Li, Gus Xia

We propose Expotion (Facial Expression and Motion Control for Multimodal Music Generation), a generative model leveraging multimodal visual controls - specifically, human facial expressions and upper-body motion - as well as text prompts to produce expressive and temporally accurate music. We adopt parameter-efficient fine-tuning (PEFT) on the pretrained text-to-music generation model, enabling fine-grained adaptation to the multimodal controls using a small dataset. To ensure precise synchronization between video and music, we introduce a temporal smoothing strategy to align multiple modalities. Experiments demonstrate that integrating visual features alongside textual descriptions enhances the overall quality of generated music in terms of musicality, creativity, beat-tempo consistency, temporal alignment with the video, and text adherence, surpassing both proposed baselines and existing state-of-the-art video-to-music generation models. Additionally, we introduce a novel dataset consisting of 7 hours of synchronized video recordings capturing expressive facial and upper-body gestures aligned with corresponding music, providing significant potential for future research in multimodal and interactive music generation.

Qi He, Gus Xia, Ziyu Wang

Hierarchical planning is a powerful approach to model long sequences structurally. Aside from considering hierarchies in the temporal structure of music, this paper explores an even more important aspect: concept hierarchy, which involves generating music ideas, transforming them, and ultimately organizing them--across musical time and space--into a complete composition. To this end, we introduce TOMI (Transforming and Organizing Music Ideas) as a novel approach in deep music generation and develop a TOMI-based model via instruction-tuned foundation LLM. Formally, we represent a multi-track composition process via a sparse, four-dimensional space characterized by clips (short audio or MIDI segments), sections (temporal positions), tracks (instrument layers), and transformations (elaboration methods). Our model is capable of generating multi-track electronic music with full-song structure, and we further integrate the TOMI-based model with the REAPER digital audio workstation, enabling interactive human-AI co-creation. Experimental results demonstrate that our approach produces higher-quality electronic music with stronger structural coherence compared to baselines.

Daniel Chin, Gus Xia

We have seen remarkable success in representation learning and language models (LMs) using deep neural networks. Many studies aim to build the underlying connections among different modalities via the alignment and mappings at the token or embedding level, but so far, most methods are very data-hungry, limiting their performance in domains such as music where paired data are less abundant. We argue that the embedding alignment is only at the surface level of multimodal alignment. In this paper, we propose a grand challenge of \textit{language model mapping} (LMM), i.e., how to map the essence implied in the LM of one domain to the LM of another domain under the assumption that LMs of different modalities are tracking the same underlying phenomena. We first introduce a basic setup of LMM, highlighting the goal to unveil a deeper aspect of cross-modal alignment as well as to achieve more sample-efficiency learning. We then discuss why music is an ideal domain in which to conduct LMM research. After that, we connect LMM in music with a more general and challenging scientific problem of \textit{learning to take actions based on both sensory input and abstract symbols}, and in the end, present an advanced version of the challenge problem setup.

Nick Bryan-Kinns, Corey Ford, Shuoyang Zheng et al.

This second international workshop on explainable AI for the Arts (XAIxArts) brought together a community of researchers in HCI, Interaction Design, AI, explainable AI (XAI), and digital arts to explore the role of XAI for the Arts. Workshop held at the 16th ACM Conference on Creativity and Cognition (C&C 2024), Chicago, USA.

Yizhi Li, Ruibin Yuan, Ge Zhang et al.

Self-supervised learning (SSL) has recently emerged as a promising paradigm for training generalisable models on large-scale data in the fields of vision, text, and speech. Although SSL has been proven effective in speech and audio, its application to music audio has yet to be thoroughly explored. This is partially due to the distinctive challenges associated with modelling musical knowledge, particularly tonal and pitched characteristics of music. To address this research gap, we propose an acoustic Music undERstanding model with large-scale self-supervised Training (MERT), which incorporates teacher models to provide pseudo labels in the masked language modelling (MLM) style acoustic pre-training. In our exploration, we identified an effective combination of teacher models, which outperforms conventional speech and audio approaches in terms of performance. This combination includes an acoustic teacher based on Residual Vector Quantisation - Variational AutoEncoder (RVQ-VAE) and a musical teacher based on the Constant-Q Transform (CQT). Furthermore, we explore a wide range of settings to overcome the instability in acoustic language model pre-training, which allows our designed paradigm to scale from 95M to 330M parameters. Experimental results indicate that our model can generalise and perform well on 14 music understanding tasks and attain state-of-the-art (SOTA) overall scores.

Qisheng Liao, Gus Xia, Zhinuo Wang

In this paper, we propose Calliffusion, a system for generating high-quality Chinese calligraphy using diffusion models. Our model architecture is based on DDPM (Denoising Diffusion Probabilistic Models), and it is capable of generating common characters in five different scripts and mimicking the styles of famous calligraphers. Experiments demonstrate that our model can generate calligraphy that is difficult to distinguish from real artworks and that our controls for characters, scripts, and styles are effective. Moreover, we demonstrate one-shot transfer learning, using LoRA (Low-Rank Adaptation) to transfer Chinese calligraphy art styles to unseen characters and even out-of-domain symbols such as English letters and digits.

Junyan Jiang, Gus Xia

We propose a novel method to model hierarchical metrical structures for both symbolic music and audio signals in a self-supervised manner with minimal domain knowledge. The model trains and inferences on beat-aligned music signals and predicts an 8-layer hierarchical metrical tree from beat, measure to the section level. The training procedure does not require any hierarchical metrical labeling except for beats, purely relying on the nature of metrical regularity and inter-voice consistency as inductive biases. We show in experiments that the method achieves comparable performance with supervised baselines on multiple metrical structure analysis tasks on both symbolic music and audio signals. All demos, source code and pre-trained models are publicly available on GitHub.

Shiqi Wei, Gus Xia

Learning symbolic music representations, especially disentangled representations with probabilistic interpretations, has been shown to benefit both music understanding and generation. However, most models are only applicable to short-term music, while learning long-term music representations remains a challenging task. We have seen several studies attempting to learn hierarchical representations directly in an end-to-end manner, but these models have not been able to achieve the desired results and the training process is not stable. In this paper, we propose a novel approach to learn long-term symbolic music representations through contextual constraints. First, we use contrastive learning to pre-train a long-term representation by constraining its difference from the short-term representation (extracted by an off-the-shelf model). Then, we fine-tune the long-term representation by a hierarchical prediction model such that a good long-term representation (e.g., an 8-bar representation) can reconstruct the corresponding short-term ones (e.g., the 2-bar representations within the 8-bar range). Experiments show that our method stabilizes the training and the fine-tuning steps. In addition, the designed contextual constraints benefit both reconstruction and disentanglement, significantly outperforming the baselines.

Ziyue Piao, Gus Xia

Breath is a significant component in singing performance, which is still underresearched in most singing-related music interfaces. In this paper, we present a multimodal system that detects the learner's singing pitch and breathing states and provides real-time visual tutoring feedback. Specifically, the breath detector is a wearable belt with pressure sensors and flexible fabric. It monitors real-time body movement of the abdomen, back waist, and twin ribs. A breath visualization algorithm is developed to display real-time breath states, together with the singing pitch contours on an interactive score interface. User studies show that our system can help users not only gain deeper breath during singing but also improve pitch accuracy in vocal training, especially for those with some musical background.

Ziyu Wang, Dejing Xu, Gus Xia et al.

Could we automatically derive the score of a piano accompaniment based on the audio of a pop song? This is the audio-to-symbolic arrangement problem we tackle in this paper. A good arrangement model should not only consider the audio content but also have prior knowledge of piano composition (so that the generation "sounds like" the audio and meanwhile maintains musicality). To this end, we contribute a cross-modal representation-learning model, which 1) extracts chord and melodic information from the audio, and 2) learns texture representation from both audio and a corrupted ground truth arrangement. We further introduce a tailored training strategy that gradually shifts the source of texture information from corrupted score to audio. In the end, the score-based texture posterior is reduced to a standard normal distribution, and only audio is needed for inference. Experiments show that our model captures major audio information and outperforms baselines in generation quality.

Jingwei Zhao, Gus Xia

Accompaniment arrangement is a difficult music generation task involving intertwined constraints of melody, harmony, texture, and music structure. Existing models are not yet able to capture all these constraints effectively, especially for long-term music generation. To address this problem, we propose AccoMontage, an accompaniment arrangement system for whole pieces of music through unifying phrase selection and neural style transfer. We focus on generating piano accompaniments for folk/pop songs based on a lead sheet (i.e., melody with chord progression). Specifically, AccoMontage first retrieves phrase montages from a database while recombining them structurally using dynamic programming. Second, chords of the retrieved phrases are manipulated to match the lead sheet via style transfer. Lastly, the system offers controls over the generation process. In contrast to pure learning-based approaches, AccoMontage introduces a novel hybrid pathway, in which rule-based optimization and deep learning are both leveraged to complement each other for high-quality generation. Experiments show that our model generates well-structured accompaniment with delicate texture, significantly outperforming the baselines.

Liwei Lin, Qiuqiang Kong, Junyan Jiang et al.

We propose a unified model for three inter-related tasks: 1) to \textit{separate} individual sound sources from a mixed music audio, 2) to \textit{transcribe} each sound source to MIDI notes, and 3) to\textit{ synthesize} new pieces based on the timbre of separated sources. The model is inspired by the fact that when humans listen to music, our minds can not only separate the sounds of different instruments, but also at the same time perceive high-level representations such as score and timbre. To mirror such capability computationally, we designed a pitch-timbre disentanglement module based on a popular encoder-decoder neural architecture for source separation. The key inductive biases are vector-quantization for pitch representation and pitch-transformation invariant for timbre representation. In addition, we adopted a query-by-example method to achieve \textit{zero-shot} learning, i.e., the model is capable of doing source separation, transcription, and synthesis for \textit{unseen} instruments. The current design focuses on audio mixtures of two monophonic instruments. Experimental results show that our model outperforms existing multi-task baselines, and the transcribed score serves as a powerful auxiliary for separation tasks.

Daniel Chin, Gus Xia

We propose the Siamese-flute method that measures the breath pressure and the acoustic sound in parallel. We fit a 6-DoF model to describe how the breath pressure affects the octave and the microtonal pitch bend, revealing the octave hysteresis. We release both our model parameters and our data analysis tools.

Ziyu Wang, Ke Chen, Junyan Jiang et al.

Music arrangement generation is a subtask of automatic music generation, which involves reconstructing and re-conceptualizing a piece with new compositional techniques. Such a generation process inevitably requires reference from the original melody, chord progression, or other structural information. Despite some promising models for arrangement, they lack more refined data to achieve better evaluations and more practical results. In this paper, we propose POP909, a dataset which contains multiple versions of the piano arrangements of 909 popular songs created by professional musicians. The main body of the dataset contains the vocal melody, the lead instrument melody, and the piano accompaniment for each song in MIDI format, which are aligned to the original audio files. Furthermore, we provide the annotations of tempo, beat, key, and chords, where the tempo curves are hand-labeled and others are done by MIR algorithms. Finally, we conduct several baseline experiments with this dataset using standard deep music generation algorithms.

Ziyu Wang, Dingsu Wang, Yixiao Zhang et al.

While deep generative models have become the leading methods for algorithmic composition, it remains a challenging problem to control the generation process because the latent variables of most deep-learning models lack good interpretability. Inspired by the content-style disentanglement idea, we design a novel architecture, under the VAE framework, that effectively learns two interpretable latent factors of polyphonic music: chord and texture. The current model focuses on learning 8-beat long piano composition segments. We show that such chord-texture disentanglement provides a controllable generation pathway leading to a wide spectrum of applications, including compositional style transfer, texture variation, and accompaniment arrangement. Both objective and subjective evaluations show that our method achieves a successful disentanglement and high quality controlled music generation.

Ziyu Wang, Yiyi Zhang, Yixiao Zhang et al.

The dominant approach for music representation learning involves the deep unsupervised model family variational autoencoder (VAE). However, most, if not all, viable attempts on this problem have largely been limited to monophonic music. Normally composed of richer modality and more complex musical structures, the polyphonic counterpart has yet to be addressed in the context of music representation learning. In this work, we propose the PianoTree VAE, a novel tree-structure extension upon VAE aiming to fit the polyphonic music learning. The experiments prove the validity of the PianoTree VAE via (i)-semantically meaningful latent code for polyphonic segments; (ii)-more satisfiable reconstruction aside of decent geometry learned in the latent space; (iii)-this model's benefits to the variety of the downstream music generation.

Tongyu Lu, Lyucheng Yan, Gus Xia

This paper proposes a word representation strategy for rhythm patterns. Using 1034 pieces of Nottingham Dataset, a rhythm word dictionary whose size is 450 (without control tokens) is generated. BERT model is created to explore syntactic potentials of rhythm words. Our model is able to find overall music structures and cluster different meters. In a larger scheme, a think mode - music as language - is proposed for systematic considerations.

Ke Chen, Gus Xia, Shlomo Dubnov

Automatic music generation is an interdisciplinary research topic that combines computational creativity and semantic analysis of music to create automatic machine improvisations. An important property of such a system is allowing the user to specify conditions and desired properties of the generated music. In this paper we designed a model for composing melodies given a user specified symbolic scenario combined with a previous music context. We add manual labeled vectors denoting external music quality in terms of chord function that provides a low dimensional representation of the harmonic tension and resolution. 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. The model contains two stages and requires separate training where the first stage adopts a Conditional Variational Autoencoder (C-VAE) to build a bijection between note sequences and their latent representations, and the second stage adopts long short-term memory networks (LSTM) with structural conditions to continue writing future melodies. We further exploit the disentanglement technique via C-VAE to allow melody generation based on pitch contour information separately from conditioning on rhythm patterns. Finally, we evaluate the proposed model using quantitative analysis of rhythm and the subjective listening study. Results show that the music generated by our model tends to have salient repetition structures, rich motives, and stable rhythm patterns. The ability to generate longer and more structural phrases from disentangled representations combined with semantic scenario specification conditions shows a broad application of our model.

Ruihan Yang, Dingsu Wang, Ziyu Wang et al.

Analogy-making is a key method for computer algorithms to generate both natural and creative music pieces. In general, an analogy is made by partially transferring the music abstractions, i.e., high-level representations and their relationships, from one piece to another; however, this procedure requires disentangling music representations, which usually takes little effort for musicians but is non-trivial for computers. Three sub-problems arise: extracting latent representations from the observation, disentangling the representations so that each part has a unique semantic interpretation, and mapping the latent representations back to actual music. In this paper, we contribute an explicitly-constrained variational autoencoder (EC$^2$-VAE) as a unified solution to all three sub-problems. We focus on disentangling the pitch and rhythm representations of 8-beat music clips conditioned on chords. In producing music analogies, this model helps us to realize the imaginary situation of "what if" a piece is composed using a different pitch contour, rhythm pattern, or chord progression by borrowing the representations from other pieces. Finally, we validate the proposed disentanglement method using objective measurements and evaluate the analogy examples by a subjective study.

Yian Zhang, Yinmiao Li, Daniel Chin et al.

Haptic interfaces have untapped the sense of touch to assist multimodal music learning. We have recently seen various improvements of interface design on tactile feedback and force guidance aiming to make instrument learning more effective. However, most interfaces are still quite static; they cannot yet sense the learning progress and adjust the tutoring strategy accordingly. To solve this problem, we contribute an adaptive haptic interface based on the latest design of haptic flute. We first adopted a clutch mechanism to enable the interface to turn on and off the haptic control flexibly in real time. The interactive tutor is then able to follow human performances and apply the "teacher force" only when the software instructs so. Finally, we incorporated the adaptive interface with a step-by-step dynamic learning strategy. Experimental results showed that dynamic learning dramatically outperforms static learning, which boosts the learning rate by 45.3% and shrinks the forgetting chance by 86%.

Ruihan Yang, Tianyao Chen, Yiyi Zhang et al.

Variational Autoencoders(VAEs) have already achieved great results on image generation and recently made promising progress on music generation. However, the generation process is still quite difficult to control in the sense that the learned latent representations lack meaningful music semantics. It would be much more useful if people can modify certain music features, such as rhythm and pitch contour, via latent representations to test different composition ideas. In this paper, we propose a new method to inspect the pitch and rhythm interpretations of the latent representations and we name it disentanglement by augmentation. Based on the interpretable representations, an intuitive graphical user interface is designed for users to better direct the music creation process by manipulating the pitch contours and rhythmic complexity.

Ziyu Wang, Gus Xia

We contribute a pop-song automation framework for lead melody generation and accompaniment arrangement. The framework reflects the major procedures of human music composition, generating both lead melody and piano accompaniment by a unified strategy. Specifically, we take chord progression as an input and propose three models to generate a structured melody with piano accompaniment textures. First, the harmony alternation model transforms a raw input chord progression to an altered one to better fit the specified music style. Second, the melody generation model generates the lead melody and other voices (melody lines) of the accompaniment using seasonal ARMA (Autoregressive Moving Average) processes. Third, the melody integration model integrates melody lines (voices) together as the final piano accompaniment. We evaluate the proposed framework using subjective listening tests. Experimental results show that the generated melodies are rated significantly higher than the ones generated by bi-directional LSTM, and our accompaniment arrangement result is comparable with a state-of-the-art commercial software, Band in a Box.

Ke Chen, Weilin Zhang, Shlomo Dubnov et al.

With recent breakthroughs in artificial neural networks, deep generative models have become one of the leading techniques for computational creativity. Despite very promising progress on image and short sequence generation, symbolic music generation remains a challenging problem since the structure of compositions are usually complicated. In this study, we attempt to solve the melody generation problem constrained by the given chord progression. This music meta-creation problem can also be incorporated into a plan recognition system with user inputs and predictive structural outputs. In particular, we explore the effect of explicit architectural encoding of musical structure via comparing two sequential generative models: LSTM (a type of RNN) and WaveNet (dilated temporal-CNN). As far as we know, this is the first study of applying WaveNet to symbolic music generation, as well as the first systematic comparison between temporal-CNN and RNN for music generation. We conduct a survey for evaluation in our generations and implemented Variable Markov Oracle in music pattern discovery. Experimental results show that to encode structure more explicitly using a stack of dilated convolution layers improved the performance significantly, and a global encoding of underlying chord progression into the generation procedure gains even more.