Recent advances in deep neural networks have enabled algorithms to compose music that is comparable to music composed by humans. However, few algorithms allow the user to generate music with tunable parameters. The ability to tune properties of generated music will yield more practical benefits for aiding artists, filmmakers, and composers in their creative tasks. In this paper, we introduce DeepJ - an end-to-end generative model that is capable of composing music conditioned on a specific mixture of composer styles. Our innovations include methods to learn musical style and music dynamics. We use our model to demonstrate a simple technique for controlling the style of generated music as a proof of concept. Evaluation of our model using human raters shows that we have improved over the Biaxial LSTM approach.
Modelling human perception of musical similarity is critical for the evaluation of generative music systems, musicological research, and many Music Information Retrieval tasks. Although human similarity judgments are the gold standard, computational analysis is often preferable, since results are often easier to reproduce, and computational methods are much more scalable. Moreover, computation based approaches can be calculated quickly and on demand, which is a prerequisite for use with an online system. We propose StyleRank, a method to measure the similarity between a MIDI file and an arbitr…
Microsoft Development Co., Ltd.（微软开发有限公司）；University of Science and Technology of China（中国科学技术大学）；Communication University of China（中国传媒大学）
Current state-of-the-art AI based classical music creation algorithms such as Music Transformer are trained by employing single sequence of notes with time-shifts. The major drawback of absolute time interval expression is the difficulty of similarity computing of notes that share the same note value yet different tempos, in one or among MIDI files. In addition, the usage of single sequence restricts the model to separately and effectively learn music information such as harmony and rhythm. In this paper, we propose a framework with two novel methods to respectively track these two shortages, …
We introduce MIDI-VAE, a neural network model based on Variational Autoencoders that is capable of handling polyphonic music with multiple instrument tracks, as well as modeling the dynamics of music by incorporating note durations and velocities. We show that MIDI-VAE can perform style transfer on symbolic music by automatically changing pitches, dynamics and instruments of a music piece from, e.g., a Classical to a Jazz style. We evaluate the efficacy of the style transfer by training separate style validation classifiers. Our model can also interpolate between short pieces of music, produce…
Télécom Paris（巴黎电信学院）；Institut Polytechnique de Paris（巴黎综合理工学院）
Research on style transfer and domain translation has clearly demonstrated the ability of deep learning-based algorithms to manipulate images in terms of artistic style. More recently, several attempts have been made to extend such approaches to music (both symbolic and audio) in order to enable transforming musical style in a similar manner. In this study, we focus on symbolic music with the goal of altering the 'style' of a piece while keeping its original 'content'. As opposed to the current methods, which are inherently restricted to be unsupervised due to the lack of 'aligned' data (i.e. …
From Generality to Mastery: Composer-Style Symbolic Music Generation via Large-Scale Pre-training
Department of Mathematics University of California, San Diego（数学系加州大学圣地亚哥分校）；Department of Computer Science and Engineering University of California, San Diego（计算机科学与工程系加州大学圣地亚哥分校）
Academia Sinica（中央研究院）；National Taiwan University（国立台湾大学）；Taiwan AI Labs（台湾人工智能实验室）
This paper proposes a new benchmark task for generat-ing musical passages in the audio domain by using thedrum loops from the FreeSound Loop Dataset, which arepublicly re-distributable. Moreover, we use a larger col-lection of drum loops from Looperman to establish fourmodel-based objective metrics for evaluation, releasingthese metrics as a library for quantifying and facilitatingthe progress of musical audio generation. Under this eval-uation framework, we benchmark the performance of threerecent deep generative adversarial network (GAN) mod-els we customize to generate loops, including Styl…
Regional style in Chinese folk songs is a rich treasure that can be used for ethnic music creation and folk culture research. In this paper, we propose MG-VAE, a music generative model based on VAE (Variational Auto-Encoder) that is capable of capturing specific music style and generating novel tunes for Chinese folk songs (Min Ge) in a manipulatable way. Specifically, we disentangle the latent space of VAE into four parts in an adversarial training way to control the information of pitch and rhythm sequence, as well as of music style and content. In detail, two classifiers are used to separat…
Carnegie Mellon University（卡内基梅隆大学）；National University of Singapore（新加坡国立大学）
Many practices have been presented in music generation recently. While stylistic music generation using deep learning techniques has became the main stream, these models still struggle to generate music with high musicality, different levels of music structure, and controllability. In addition, more application scenarios such as music therapy require imitating more specific musical styles from a few given music examples, rather than capturing the overall genre style of a large data corpus. To address requirements that challenge current deep learning methods, we propose a statistical machine le…
We consider the problem of learning high-level controls over the global structure of generated sequences, particularly in the context of symbolic music generation with complex language models. In this work, we present the Transformer autoencoder, which aggregates encodings of the input data across time to obtain a global representation of style from a given performance. We show it is possible to combine this global representation with other temporally distributed embeddings, enabling improved control over the separate aspects of performance style and melody. Empirically, we demonstrate the eff…
The subjective evaluation of music generation techniques has been mostly done with questionnaire-based listening tests while ignoring the perspectives from music composition, arrangement, and soundtrack editing. In this paper, we propose an editing test to evaluate users' editing experience of music generation models in a systematic way. To do this, we design a new music style transfer model combining the non-chronological inference architecture, autoregressive models and the Transformer, which serves as an improvement from the baseline model on the same style transfer task. Then, we compare t…
ImprovNet -- Generating Controllable Musical Improvisations with Iterative Corruption Refinement
Electronic Engineering and Computer Science Department, Queen Mary University of London, UK（伦敦大学女王学院电子工程与计算机科学系）；Information Systems, Technology, and Design, Singapore University of Technology and Design, Singapore（新加坡科技设计大学信息系统、科技与设计系）
Music and Audio Research Group (MARG), Seoul National University（首尔国立大学音乐与音频研究组（MARG））；supertone Inc.（supertone公司）
This paper proposes a controllable singing voice synthesis system capable of generating expressive singing voice with two novel methodologies. First, a local style token module, which predicts frame-level style tokens from an input pitch and text sequence, is proposed to allow the singing voice system to control musical expression often unspecified in sheet music (e.g., breathing and intensity). Second, we propose a dual-path pitch encoder with a choice of two different pitch inputs: MIDI pitch sequence or f0 contour. Because the initial generation of a singing voice is usually executed by tak…
Taiwan AI Labs（台湾人工智能实验室）；National Taiwan University（台湾大学）；Academia Sinica（中央研究院）
Transformers and variational autoencoders (VAE) have been extensively employed for symbolic (e.g., MIDI) domain music generation. While the former boast an impressive capability in modeling long sequences, the latter allow users to willingly exert control over different parts (e.g., bars) of the music to be generated. In this paper, we are interested in bringing the two together to construct a single model that exhibits both strengths. The task is split into two steps. First, we equip Transformer decoders with the ability to accept segment-level, time-varying conditions during sequence generat…
Q&A: Query-Based Representation Learning for Multi-Track Symbolic Music re-Arrangement
Institute of Data Science, NUS（新加坡国立大学数据科学研究所）；Integrative Sciences and Engineering Programme, NUS Graduate School（新加坡国立大学研究生院综合科学与工程项目）；Music X Lab, NYU Shanghai（上海纽约大学Music X实验室）；MBZUAI（穆罕默德·本·扎耶德人工智能大学）；School of Computing, NUS（新加坡国立大学计算学院）
Performance Conditioning for Diffusion-Based Multi-Instrument Music Synthesis
International Audio Laboratories Erlangen（埃尔朗根国际音频实验室）；Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡弗里德里希-亚历山大大学）；Fraunhofer Institute for Integrated Circuits IIS（弗劳恩霍夫集成电路研究所IIS）
Sony Computer Science Laboratories, Paris（索尼计算机科学实验室（巴黎））；LIP6, Université Pierre et Marie Curie（皮埃尔和玛丽·居里大学LIP6实验室）
Modeling polyphonic music is a particularly challenging task because of the intricate interplay between melody and harmony. A good model should satisfy three requirements: statistical accuracy (capturing faithfully the statistics of correlations at various ranges, horizontally and vertically), flexibility (coping with arbitrary user constraints), and generalization capacity (inventing new material, while staying in the style of the training corpus). Models proposed so far fail on at least one of these requirements. We propose a statistical model of polyphonic music, based on the maximum entrop…
We address the challenging open problem of learning an effective latent space for symbolic music data in generative music modeling. We focus on leveraging adversarial regularization as a flexible and natural mean to imbue variational autoencoders with context information concerning music genre and style. Through the paper, we show how Gaussian mixtures taking into account music metadata information can be used as an effective prior for the autoencoder latent space, introducing the first Music Adversarial Autoencoder (MusAE). The empirical analysis on a large scale benchmark shows that our mode…
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 ret…
Symbolic Music Generation relies on the contextual representation capabilities of the generative model, where the most prevalent approach is the Transformer-based model. The learning of musical context is also related to the structural elements in music, i.e. intro, verse, and chorus, which are currently overlooked by the research community. In this paper, we propose a hierarchical Transformer model to learn multi-scale contexts in music. In the encoding phase, we first designed a Fragment Scope Localization layer to syncopate the music into chords and sections. Then, we use a multi-scale atte…
Northwestern University（西北大学）；Bienen School of Music, Northwestern University（西北大学比嫩音乐学院）；Sony CSL（索尼计算机科学实验室）
Deep learning has rapidly become the state-of-the-art approach for music generation. However, training a deep model typically requires a large training set, which is often not available for specific musical styles. In this paper, we present augmentative generation (Aug-Gen), a method of dataset augmentation for any music generation system trained on a resource-constrained domain. The key intuition of this method is that the training data for a generative system can be augmented by examples the system produces during the course of training, provided these examples are of sufficiently high quali…
Deep generative models such as Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) have recently been applied to style and domain transfer for images, and in the case of VAEs, music. GAN-based models employing several generators and some form of cycle consistency loss have been among the most successful for image domain transfer. In this paper we apply such a model to symbolic music and show the feasibility of our approach for music genre transfer. Evaluations using separate genre classifiers show that the style transfer works well. In order to improve the fidelity of th…
From West to East: Who can understand the music of the others better?
National Technical University of Athens（雅典国立技术大学）；School of ECE, National Technical University of Athens（雅典国立技术大学电子与计算机工程学院）；Queen Mary University of London（伦敦大学玛丽皇后学院）；Centre for Digital Music, Queen Mary University of London（伦敦大学玛丽皇后学院数字音乐中心）
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 g…
LIP6, Université Pierre et Marie Curie（皮埃尔和玛丽·居里大学 LIP6 实验室）；Sony CSL, Paris（索尼计算机科学实验室（巴黎））；École Polytechnique, Palaiseau, France（巴黎综合理工学院）；Sony CSL, Tokyo（索尼计算机科学实验室（东京））
Distances on symbolic musical sequences are needed for a variety of applications, from music retrieval to automatic music generation. These musical sequences belong to a given corpus (or style) and it is obvious that a good distance on musical sequences should take this information into account; being able to define a distance ex nihilo which could be applicable to all music styles seems implausible. A distance could also be invariant under some transformations, such as transpositions, so that it can be used as a distance between musical motives rather than musical sequences. However, to our k…
Queen Mary University of London（伦敦玛丽女王大学）；Academia Sinica（中央研究院）；Indian Institute of Technology Kharagpur（印度理工学院卡拉格普尔分校）；Dadabots；Taiwan AI Labs（台湾人工智能实验室）
Originating in the Renaissance and burgeoning in the digital era, tablatures are a commonly used music notation system which provides explicit representations of instrument fingerings rather than pitches. GuitarPro has established itself as a widely used tablature format and software enabling musicians to edit and share songs for musical practice, learning, and composition. In this work, we present DadaGP, a new symbolic music dataset comprising 26,181 song scores in the GuitarPro format covering 739 musical genres, along with an accompanying tokenized format well-suited for generative sequenc…
Structured Multi-Track Accompaniment Arrangement via Style Prior Modelling
Institute of Data Science, NUS（新加坡国立大学数据科学研究院）；Integrative Sciences and Engineering Programme, NUS Graduate School（新加坡国立大学研究生院整合科学与工程项目）；Machine Learning Department, MBZUAI（穆罕默德·本·扎耶德人工智能大学机器学习系）；Computer Science Department, NYU Shanghai（上海纽约大学计算机科学系）；School of Computing, NUS（新加坡国立大学计算机学院）
MusicMamba: A Dual-Feature Modeling Approach for Generating Chinese Traditional Music with Modal Precision
School of Computer Science and Artificial Intelligence, Wuhan University of Technology（武汉理工大学计算机科学与人工智能学院）；School of Computer Science, Hubei University of Technology（湖北工业大学计算机学院）
Google Research（谷歌研究院）；IRCAM - Sorbonne Université（巴黎声学音乐研究协调所-索邦大学）
We introduce MusicLM, a model generating high-fidelity music from text descriptions such as "a calming violin melody backed by a distorted guitar riff". MusicLM casts the process of conditional music generation as a hierarchical sequence-to-sequence modeling task, and it generates music at 24 kHz that remains consistent over several minutes. Our experiments show that MusicLM outperforms previous systems both in audio quality and adherence to the text description. Moreover, we demonstrate that MusicLM can be conditioned on both text and a melody in that it can transform whistled and hummed melo…
In this paper we introduce a novel feature augmentation approach for generating structured musical compositions comprising melodies and harmonies. The proposed method augments a connectionist generation model with count-down to song conclusion and meter markers as extra input features to study whether neural networks can learn to produce more aesthetically pleasing and structured musical output as a consequence of augmenting the input data with structural features. An RNN architecture with LSTM cells is trained on the Nottingham folk music dataset in a supervised sequence learning setup, follo…
Discovering "Words" in Music: Unsupervised Learning of Compositional Sparse Code for Symbolic Music
Central Conservatory of Music（中央音乐学院）；Department of Industrial Engineering, Ajou University（Ajou大学工业工程系）；Alibaba Group（阿里巴巴集团）；Bigai；Peking University（北京大学）
While deep generative models have empowered music generation, it remains a challenging and under-explored problem to edit an existing musical piece at fine granularity. In this paper, we propose SDMuse, a unified Stochastic Differential Music editing and generation framework, which can not only compose a whole musical piece from scratch, but also modify existing musical pieces in many ways, such as combination, continuation, inpainting, and style transferring. The proposed SDMuse follows a two-stage pipeline to achieve music generation and editing on top of a hybrid representation including pi…
Indian Institute of Information Technology, Sri City（斯里城印度信息技术学院）
Realistic music generation has always remained as a challenging problem as it may lack structure or rationality. In this work, we propose a deep learning based music generation method in order to produce old style music particularly JAZZ with rehashed melodic structures utilizing a Bi-directional Long Short Term Memory (Bi-LSTM) Neural Network with Attention. Owing to the success in modelling long-term temporal dependencies in sequential data and its success in case of videos, Bi-LSTMs with attention serve as the natural choice and early utilization in music generation. We validate in our expe…
NYU Shanghai（上海纽约大学）；Queen Mary University of London（伦敦玛丽女王大学）；Tencent Inc.（腾讯公司）；MBZUAI（穆罕默德·本·扎耶德人工智能大学）
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 …
This paper presents an architecture for generating music for video games based on the Transformer deep learning model. Our motivation is to be able to customize the generation according to the taste of the player, who can select a corpus of training examples, corresponding to his preferred musical style. The system generates various musical layers, following the standard layering strategy currently used by composers designing video game music. To adapt the music generated to the game play and to the player(s) situation, we are using an arousal-valence model of emotions, in order to control the…
Despite the central role that melody plays in music perception, it remains an open challenge in music information retrieval to reliably detect the notes of the melody present in an arbitrary music recording. A key challenge in melody transcription is building methods which can handle broad audio containing any number of instrument ensembles and musical styles - existing strategies work well for some melody instruments or styles but not all. To confront this challenge, we leverage representations from Jukebox (Dhariwal et al. 2020), a generative model of broad music audio, thereby improving per…
Peking University（北京大学）；New York University Shanghai（上海纽约大学）
Led by the success of neural style transfer on visual arts, there has been a rising trend very recently in the effort of music style transfer. However, "music style" is not yet a well-defined concept from a scientific point of view. The difficulty lies in the intrinsic multi-level and multi-modal character of music representation (which is very different from image representation). As a result, depending on their interpretation of "music style", current studies under the category of "music style transfer", are actually solving completely different problems that belong to a variety of sub-field…
