Music Generation Research Articles

Personalized Music Playlist Generation Method Based on Personal Impression and Pleasure

Personalized Music Playlist Generation Method Based on Personal Impression and Pleasure

Personalized Music Playlist Generation Method for Placing the Listener in a Positive Mood

Personalized Music Playlist Generation Method for Placing the Listener in a Positive Mood

Symbolic Music Generation Conditioned on Continuous-Valued Emotions

In this paper we present a new approach for the generation of multi-instrument symbolic music driven by musical emotion. The principal novelty of our approach centres on conditioning a state-of-the-art transformer based on continuous-valued valence and arousal labels. In addition, we provide a new large-scale dataset of symbolic music paired with emotion labels in terms of valence and arousal. We evaluate our approach in a quantitative manner in two ways, first by measuring its note prediction accuracy, and second via a regression task in the valence-arousal plane. Our results demonstrate that our proposed approaches outperform conditioning using control tokens which is representative of the current state of the art.

Use of Speaker Recognition Approaches for Learning and Evaluating Embedding Representations of Musical Instrument Sounds

Constructing an embedding space for musical instrument sounds that can meaningfully represent new and unseen instruments is important for downstream music generation tasks such as multi-instrument synthesis and timbre transfer. The framework of Automatic Speaker Verification (ASV) provides us with architectures and evaluation methodologies for verifying the identities of unseen speakers, and these can be repurposed for the task of learning and evaluating a musical instrument sound embedding space that can support unseen instruments. Borrowing from state-of-the-art ASV techniques, we construct a musical instrument recognition model that uses a SincNet front-end, a ResNet architecture, and an angular softmax objective function. Experiments on the NSynth and RWC datasets show our model’s effectiveness in terms of equal error rate (EER) for unseen instruments, and ablation studies show the importance of data augmentation and the angular softmax objective. Experiments also show the benefit of using a CQT-based filterbank for initializing SincNet over a Mel filterbank initialization. Further complementary analysis of the learned embedding space is conducted with t-SNE visualizations and probing classification tasks, which show that including instrument family labels as a multi-task learning target can help to regularize the embedding space and incorporate useful structure, and that meaningful information such as playing style, which was not included during training, is contained in the embeddings of unseen instruments.

Research and Implementation of Automatic Composition System Based on ACMN

The use of artificial neural network to compose music is a new computer composition method proposed in recent years. The advantage of this method is that it can process large-scale data in a relatively short time, greatly reducing the preparation work before computer composition and improving composition efficiency. However, there are still many problems that have not made obvious breakthroughs. For example, the yield is low, and a considerable proportion of the generated works have no appreciation value. They are completely random patchwork of notes and chords in the time dimension; the music produced is not pleasant, the melody is too simple and so on. In order to make the songs generated by the composition system more in line with the rules of music theory, this paper proposes a music generation model ACMN (Actor-Critic Music Network) based on the policy gradient method, which is characterized by the combination of reinforcement learning technology and neural network. Experiments show that, compared with the composition model without reinforcement learning technology, the ACMN model has more outstanding performance in terms of repetition rate and pleasantness.

Композитор Азат Шишян / Composer Azat Shishyan

Composer, Azat Shishyan, an honored teacher of RA, musical-public figure is a representative of the musical generation of the second half of the 20h century, whose dedication to art and modest work earned the admiration and respect of his contemporaries. Azat Shishyan's musical legacy includes four symphonies, an opera, a ballet, chamber and choral compositions, songs and romances, music for theater performances, as well as creations for children. The composer's creative consciousness was profoundly influenced by the words of his teacher, B. Kanachyan: "Don 't forget that you are the offsprings of Komitas". Shishyan's creative journey was fundamentally grounded in embracing and embodying Komitas' artistic principles from his earliest endeavors. In addition to his composition work, Shishyan dedicated himself to teaching, nurturing the talents of renowned composers such as M. Israelyan, Y. Yerkanyan, A. Nersisyan, G. Chitchyan, and T. Mansuryan, all of whom took their initial steps under his guidance. Equally noteworthy is Shishyan 's significant role in enriching the cultural life of Leninakan through his musical and educational contributions.

MUSIC GENERATION USING DEEP LEARNING

We examine how lengthy short-term memory neural networks (NNs) may be utilised to create music compositions and offer a method for doing so in this study. Bach's musical style was chosen to train the NN in order for it to make similar music works. The recommended method converts midi files to song files before encoding them as NN inputs. Before feeding the files into the NNs, they are augmented, which converts them into distinct keys, and then they are fed into the NN for training. The final phase is the creation of music. The primary purpose is to assign an arbitrary note to the NN, which it will gradually modify until it produces a good piece of music. Several tests have been conducted in order to identify the ideal parameter values for producing good music. Keywords: lstm, music generation, deep learning, machine learning, neural networks

Algorithmic music generation by harmony recombination with genetic algorithm

Algorithmic music composition has recently become an area of prestigious research in projects such as Google’s Magenta, Aiva, and Sony’s CSL Lab aiming to increase the composers’ tools for creativity. There are advances in systems for music feature extraction and generation of harmonies with short-time and long-time patterns of music style, genre, and motif. However, there are still challenges in the creation of poly-instrumental and polyphonic music, pieces become repetitive and sometimes these systems copy the original files. The main contribution of this paper is related to the improvement of generating new non-plagiary harmonic developments constructed from the symbolic abstraction from MIDI music non-labeled data with controlled selection of rhythmic features based on evolutionary techniques. Particularly, a novel approach for generating new music compositions by replacing existing harmony descriptors in a MIDI file with new harmonic features from another MIDI file selected by a genetic algorithm. This allows combining newly created harmony with a rhythm of another composition guaranteeing the adjustment of a new music piece to a distinctive genre with regularity and consistency. The performance of the proposed approach has been assessed using artificial intelligent computational tests, which assure goodness of the extracted features and shows its quality and competitiveness.

The Ai Music Generation Challenge 2020: Double Jigs in the Style of O'Neill's ``1001''

The Ai Music Generation Challenge 2020 had three objectives: 1) to promote meaningful approaches to evaluating artificial intelligence (Ai) applied to music;2) to see how music Ai research can benefit from considering traditional music, and how traditional music might benefit from music Ai research; and 3)to facilitate discussions about the ethics of music Ai research applied to traditional music practices.There were six participants and a benchmark in the challenge, each competing to build an artificial system that generates the most plausible double jigs, as judged against the 365 published in solved'', but that the evaluation of such systems can be done in meaningful ways.The article ends by reflecting on the challenge and considering the coming 2021 challenge.

I Keep Counting: An Experiment in Human/AI Co-creative Songwriting

Musical co-creativity aims at making humans and computers collaborate to compose music. As an MIR team in computational musicology, we experimented with co-creativity when writing our entry to the “AI Song Contest 2020”. Artificial intelligence was used to generate the song’s structure, harmony, lyrics, and hook melody independently and as a basis for human composition. It was a challenge from both the creative and the technical point of view: in a very short time-frame, the team had to adapt its own simple models, or experiment with existing ones, to a related yet still unfamiliar task, music generation through AI. The song we propose is called “I Keep Counting”. We openly detail the process of songwriting, arrangement, and production. This experience raised many questions on the relationship between creativity and machine, both in music analysis and generation, and on the role AI could play to assist a composer in their work. We experimented with <em>AI as automation</em>, mechanizing some parts of the composition, and especially <em>AI as suggestion</em> to foster the composer’s creativity, thanks to surprising lyrics, uncommon successions of sections and unexpected chord progressions. Working with this material was thus a stimulus for human creativity.

The Future of Music Copyright Collectives in the Digital Streaming Age

The Future of Music Copyright Collectives in the Digital Streaming Age

MCMA: A Symbolic Multitrack Contrapuntal Music Archive

We present Multitrack Contrapuntal Music Archive (MCMA, available at https://mcma.readthedocs.io), a symbolic dataset of pieces specifically curated to comprise, for any given polyphonic work, independent voices. So far, MCMA consists only of pieces from the Baroque repertoire but we aim to extend it to other contrapuntal music. MCMA is FAIR-compliant and it is geared towards musicological tasks such as (computational) analysis or education, as it brings to the fore contrapuntal interactions by explicit and independent representation. Furthermore, it affords for a more apt usage of recent advances in the field of natural language processing (e.g., neural machine translation). For example, MCMA can be particularly useful in the context of language-based machine learning models for music generation. Despite its current modest size, we believe MCMA to be an important addition to online contrapuntal music databases, and we thus open it to contributions from the wider community, in the hope that MCMA can continue to grow beyond our efforts. In this article, we provide the rationale for this corpus, suggest possible use cases, offer an overview of the compiling process (data sourcing and processing), and present a brief statistical analysis of the corpus at the time of writing. Finally, future work that we endeavor to undertake is discussed.

MeloSol Corpus

This data report introduces the MeloSol corpus, a collection of 783 Western, tonal monophonic melodies. I first begin by describing the overall structure of the corpus, then proceed to detail its contents as they would be helpful for researchers working in the field of computational musicology or music psychology. In order to contextualize the MeloSol corpus in relation to other corpora in the literature, I present descriptive statistics of the MeloSol corpus alongside the The Densmore Collection of Native American Song and The Essen Folk Song Collection. I suggest possible future uses of this corpus including extending research investigating Western tonality, perceptual experiments needing novel ecological stimuli, or work involving the musical generation of monophonic melodies in the style of Western tonal music.

Drumroll Please: Modeling Multi-Scale Rhythmic Gestures with Flexible Grids

We present a new data representation for music modeling and generation called a Flexible Grid. This representation aims to balance flexibility with structure in order to encode all the musical events (notes or rhythmic onsets) in a dataset without quantizing or discarding any temporal information. In experiments with a dataset of MIDI drum performances, we find that when implemented in a Variational AutoEncoder (VAE) model, Flexible Grid representations can enable detailed generation of music performance data that includes multiple different gestures and articulations.

A transformer generative adversarial network for multi‐track music generation

This study proposes a new generation network based on transformers and guided by the music theory to produce high-quality music work. In this study, the decoding block of the transformer is used to learn the internal information of single-track music, and cross-track transformers are used to learn the information amongst the tracks of different musical instruments. A reward network based on the music theory is proposed, which optimizes the global and local loss objective functions while training and discriminating the network so that the reward network can provide a reliable adjustment method for the generation of the network. The method of combining the reward network and cross entropy loss is used to guide the training of the generator and produce high-quality music work. Compared with other multi-track music generation models, the experimental results verify the validity of the model.

Research on the Algorithm System for Generation and Creation by Computer Artificial Intelligence and Support Vector Machine

Algorithmic composition is also called automated composition. It is an attempt to use a specific form of process. Composers make full use of computers to carry out music creation and reduce their access. In this paper, based on the standard support vector machine (SVM) learning neural network, the least square support vector machine (LS-SVM) is combined with the recurrent neural network, and a new least square support vector machine learning neural network is proposed. The article realizes the efficient end-to-end multi-dimensional sound wave time series generation model Music-coder, through which the music style music of the famous singer Jay Chou is generated, and the quantified similarity with the real Jay Chou music data set reaches a maximum of 97.73%. The project in this paper shows that intelligent algorithm as a composition tool for music generation and creation is an effective music production program and will bring new development to music production.

Assessing Open-Ended Human-Computer Collaboration Systems: Applying a Hallmarks Approach.

There is a growing desire to create computer systems that can collaborate with humans on complex, open-ended activities. These activities typically have no set completion criteria and frequently involve multimodal communication, extensive world knowledge, creativity, and building structures or compositions through multiple steps. Because these systems differ from question and answer (Q&A) systems, chatbots, and simple task-oriented assistants, new methods for evaluating such collaborative computer systems are needed. Here, we present a set of criteria for evaluating these systems, called Hallmarks of Human-Machine Collaboration. The Hallmarks build on the success of heuristic evaluation used by the user interface community and past evaluation techniques used in the spoken language and chatbot communities. They consist of observable characteristics indicative of successful collaborative communication, grouped into eight high-level properties: robustness; habitability; mutual contribution of meaningful content; context-awareness; consistent human engagement; provision of rationale; use of elementary concepts to teach and learn new concepts; and successful collaboration. We present examples of how we used these Hallmarks in the DARPA Communicating with Computers (CwC) program to evaluate diverse activities, including story and music generation, interactive building with blocks, and exploration of molecular mechanisms in cancer. We used the Hallmarks as guides for developers and as diagnostics, assessing systems with the Hallmarks to identify strengths and opportunities for improvement using logs from user studies, surveying the human partner, third-party review of creative products, and direct tests. Informal feedback from CwC technology developers indicates that the use of the Hallmarks for program evaluation helped guide development. The Hallmarks also made it possible to identify areas of progress and major gaps in developing systems where the machine is an equal, creative partner.

Towards decolonising university music education in Nigeria

ABSTRACT This article argues that coloniality is an ongoing feature of university music education in Nigeria. It uses a multiple colonialisms framework in exploring Nigerian higher music education systems as historical and contemporary sites of colonialism within which Nigerian universities engage in music knowledge generation to reach this conclusion. It relies on information gleaned from scholarly sources. It shows that the dominance of Euro-American classical musical practice in bicultural university music education in Nigeria since 1961 has received little questioning. It identifies decolonisation as indigenisation, Africanisation, Islamisation, Nigerianisation and globalisation of university music education discourses as competing with this dominant paradigm. From the anti-colonial perspective, it argues that these six Nigerian music knowledge systems should not be simply celebrated, valorised and utilised. They should be critically engaged, interrogated, transformed, altered if necessary, and studied relationally, contextually and comparatively. Decolonising university music education in any society, it posits, involves centring music knowledge creation, understanding and sharing on multiplicity and entanglements.

Apollo: A Classical Piano Composer Using Long Short-Term Memory

With Deep Learning becoming more accessible than ever, music generation has become a domain that has piqued the interest of computer scientists and musicians’ with a scientific bent. Recent advances in Deep Learning have made it possible to compose music almost indistinguishable from those fashioned by humans. However, any musical model that proposes to produce a sequence of harmonious music necessitates that the model should have the potential to recollect past details and have a sound understanding of the musical structure. Hence, in this paper, we propose a Deep Learning architecture to generate piano symphonies by selecting a list of notes and using the power of Long Short-Term Memory (LSTM) to predict the next note based on the discography of the given composers, such as Ludwig van Beethoven and Wolfgang Amadeus Mozart.

Research on Music Copy Generation Method of Smart Speakers

Smart speakers have become more and more popular in recent years with the continuous development of artificial intelligence. In daily life, dialogues between users and smart speakers can realize music play now. Currently, most manufacturers have made smart speakers directly play the song that the user wants to listen to, or inserted a fixed-style copy before playing. To provide users with a better music experience and allow users to have a deeper understanding of songs, a radio-like approach is planned to be adopted, inserting some personalized music copy before smart speakers play music. In this article, TextRank-based automatic text summarization and keyword extraction methods are used to efficiently and automatically generate short, appropriate, and personalized music copy from the song library, and insert the copy before the smart speaker plays the song. This will not only increase the users’ music experience, but also allow users to express their emotions with music as sustenance. The research on the music copy generation technology of smart speakers can not only improve the music experience of users but also has a great impetus to improve the functions of smart speakers.