Musical Sequences Research Articles

TARREAN: A Novel Transformer with a Gate Recurrent Unit for Stylized Music Generation.

Music generation by AI algorithms like Transformer is currently a research hotspot. Existing methods often suffer from issues related to coherence and high computational costs. To address these problems, we propose a novel Transformer-based model that incorporates a gate recurrent unit with root mean square norm restriction (TARREAN). This model improves the temporal coherence of music by utilizing the gate recurrent unit (GRU), which enhances the model's ability to capture the dependencies between sequential elements. Additionally, we apply masked multi-head attention to prevent the model from accessing future information during training, preserving the causal structure of music sequences. To reduce computational overhead, we introduce root mean square layer normalization (RMS Norm), which smooths gradients and simplifies the calculations, thereby improving training efficiency. The music sequences are encoded using a compound word method, converting them into discrete symbol-event combinations for input into the TARREAN model. The proposed method effectively mitigates discontinuity issues in generated music and enhances generation quality. We evaluated the model using the Essen Associative Code and Folk Song Database, which contains 20,000 folk melodies from Germany, Poland, and China. The results show that our model produces music that is more aligned with human preferences, as indicated by subjective evaluation scores. The TARREAN model achieved a satisfaction score of 4.34, significantly higher than the 3.79 score of the Transformer-XL + REMI model. Objective evaluation also demonstrated a 15% improvement in temporal coherence compared to traditional methods. Both objective and subjective experimental results demonstrate that TARREAN can significantly improve generation coherence and reduce computational costs.

Problems of music editing and expression of the violin composition

Problems of music editing and expression of the violin composition

Masked Expressiveness: Conditioned Generation of Piano Key Striking Velocity Using Masked Language Modeling

Creating and experiencing expressive renditions of composed musical pieces are fundamental to how people engage with music. Numerous studies have focused on building computational models of musical expressiveness to gain a deeper understanding of its nature and explore potential applications. This paper examines masked language modeling (MLM) for modeling expressiveness in piano performance, specifically targeting the prediction of key striking velocity using vanilla Bidirectional Encoder Representations from Transformers (BERT). While MLM has been explored in previous studies, this work applies it in a novel direction by concentrating on the fine-grained conditioned prediction of velocity information. The results show that the model can predict masked velocity events in various contexts within an acceptable margin of error, relying solely on the pitch, timing, and velocity data encoded in Musical Instrument Digital Interface (MIDI) files. Additionally, the study employs a sequential masking and prediction approach toward rendering the velocity of unseen MIDI files and achieves more musically convincing results. This approach holds promise for developing interactive systems for expressive performance generation, such as advanced piano conducting or accompaniment systems.

Effects of a 1-year piano intervention on cognitive flexibility in older adults.

We analyzed the data of a randomized controlled trial to examine the effects of piano practice on cognitive flexibility in healthy older adults. Participants (N = 153, 69.5 ± 3.5 years of age, 57.5% females) were randomly assigned to a piano practice group (PP) or a control group engaged in active music listening (MC). Both groups underwent a year-long intervention with weekly 60-min lessons and daily homework. We assessed switch and mixing costs in terms of speed (mean reaction times) and variability (standard deviation of reaction times) with a number switch and a perceptual switch task. We employed scale analysis based on musical instrument digital interface to assess pianistic performance. Tests were conducted at baseline, after 6 months, postintervention (12 months), and at follow-up (18 months). Results revealed more pronounced improvements in pianistic performance in the PP group compared with the MC group over the course of the intervention. Both groups exhibited gains in several cognitive flexibility outcomes, which originated primarily in the latter half of the intervention. For mixing costs of the number switch test, the PP group showed greater improvements compared with the MC group. Changes in pianistic performance were not related to changes in cognitive flexibility. Additionally, the findings indicated a compensation account in both groups, which was more pronounced in the PP group for mixing costs. This study suggests that both piano practice and active music listening-with piano practice to a higher degree-enhance cognitive flexibility, particularly in sustained control mechanisms linked to mixing costs. Both interventions potentially require longer than 6 months to induce behavioral transfer effects and are especially beneficial for individuals with lower cognitive flexibility levels. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

ნიკოლოზ ბარათაშვილის ლექსის „როს ბედნიერ ვარ შენთან ყოფნითა“ პირველწყაროსათვის

In the notes and comments on the poetic works of Nikoloz Bara­tashvili, it is erroneously stated that his poem “When I am happy with your presence” is based on the anonymous Russian song "The Brook" that is suggested by Nikolai in Leo Tolstoy’s novel “War and Peace” and the text of which is quoted by the writer's wife Sophia Tolstaya in her memoirs “My Life”. But in reality, as we have managed to find out, “When I am happy with your presence” has nothing to do with “The Brook”. Nikoloz Bara­tashvili’s poem is s a translation of the once popular romance by the Russian composer Pyotr Bulakhov to the words of “Kol’ sĉastliv ja s t’aboi bivayu” (“How I am happy with you”) by Vasily Chuyevsky, a Ru­ssian poet of the mid-19th century. Nikoloz Baratashvili’s trans­lation is as close as possible to the Russian source, to its meter, into­nation and refrain in the stanzas and key images of the original source. At the same time in Georgian translation some lines are modified and the melodic tonality of the original text is intensified. The date of Nikoloz Baratashvili's translation of the Russian romance is not indicated in the autograph. According to the editors of publications of his works, the translation was made in 1841. Biographical information about Vasily Chuevsky is almost comple­tely absent. It is only reliably known that he was one of the most remarkable poets and songwriters of the mid-19th century. There is information that in the mid-1840s, while being a student at Moscow University, he, together with Alexander Dubuque, a Russian com­poser and music teacher, published musical editions of romances and songs in 1846. Taking into account the year of Nikoloz Baratashvili's death, 1845, and the fact that some time had to pass before Vasily Chuevsky's ro­mance could come to the attention of the Georgian poet, certain conclu-sions can be made about the time of the creation of the Russian romance and its adaptation into Georgian. The translation of the Russian romance by Nikoloz Barata­shvili ex­pands our understanding of the character of the develop­ment of vocal lyrics in Georgian romantic poetry and, at the same time, deserves attention because there is almost no data on the life, creative activity and lifetime popularity of Vasily Chuyevsky. We attach to the article Petr Bulakhov’s notes on the words of the romance by Vasily Chuevsky, expanding our understanding of the nature of the music and text that attracted the attention of Nikoloz Baratashvili.

Music Sequence Creation Technique in Piano Music Production Based on Embedded Neural Network

With the rapid development of computer technology and artificial intelligence, especially the rise of embedded neural network technology. The traditional arrangement methods are facing increasing challenges in the field of music production, and their complexity and time-consuming nature are no longer sufficient to meet the urgent needs of modern audiences for efficient and high-quality computer piano music works. This paper proposes an innovative solution — a piano music production matching method based on embedded neural network fusion, and particularly focused on intelligent creation technology for music sequences. In the system, embedded neural networks are core applied in the process of creating music sequences. By utilizing their powerful feature learning and generalization abilities, neural networks can capture complex patterns and dynamic changes in music sequences, providing infinite possibilities for melody generation and innovation. At the same time, combined with association rule algorithms, the system can discover potential association patterns between notes, further optimizing the fluency and harmony of the melody; The clustering algorithm helps extract common features from massive music data, providing a rich material library for melody innovation.

Stochastic properties of musical time series

Musical sequences are correlated dynamical processes that may differ depending on musical styles. We aim to quantify the correlations through power spectral analysis of pitch sequences in a large corpus of musical compositions as well as improvised performances. Using a multitaper method we extend the power spectral estimates down to the smallest possible frequencies optimizing the tradeoff between bias and variance. The power spectral densities reveal a characteristic behavior; they typically follow inverse power laws (1/f β-noise), yet only down to a cutoff frequency, where they end in a plateau. Correspondingly the pitch autocorrelation function exhibits slow power law decays only up to a cutoff time, beyond which the correlations vanish. We determine cutoff times between 4 and 100 quarter note units for the compositions and improvisations of the corpus, serving as a measure for the degree of persistence and predictability in music. The histogram of exponents β for the power law regimes has a pronounced peak near β = 1 for classical compositions, but is much broader for jazz improvisations.

‘Texting Scarlatti’: unlocking a standard edition with a digital toolkit

Editing musical texts requires a great deal of artistic, intellectual and commercial investment but, once published, an edition tends to become inert, a form of musical still life. ‘Texting Scarlatti’ explores ways of unlocking the value of a standard edition using a variety of digital tools. For as long as the editorial process in music begins and ends with analogue graphical notation, a digital music edition will inevitably be hybrid to some extent. But our project examines each stage in that process from input to output, showing how scholarly insight, performance-based musicology, data science, HCI-inspired crowdsourcing techniques and information systems can together form the basis of an open, rich digital edition for future development. Key outputs will be an extremely large machine-readable dataset consisting of more than 310,000 bars of music, together with structured, searchable apparatus for each of the 555 sonatas in the Scarlatti canon. Musicology has been a notable absentee from recent developments in artificial intelligence and machine learning (AIML) and there is a great deal of catching-up to do. In our conclusion, we urge more scholars and publishers to make similarly large, and largely hidden, datasets more openly available for machine learning and suggest some practical steps to facilitate this.

Generative Adversarial Network Based Music Generation

Music has been an integral part of human civilization personally and culturally. Historically, music has been generated using various instruments, or natural sounds like water drops, or unconventional musical instruments like metal or glass-wares. At present, technologies like Musical Instrument Digital Interface (MIDI) are used to generate music electronically. This research investigates the use of Generative Adversarial Networks (GANs) for beginner-friendly music production. This model uses Long Short-Term Memory (LSTM) generator and Patch GAN as discriminator for the GAN architecture. The generator consists of input layer, embedding layer, LSTM layer and generates output with a SoftMax function. Similarly, the discriminator consists of a convolution layer, the output of which is averaged by the global average pooling layer, and output is generated by the sigmoid function. The model is trained on Maestro MIDI dataset. We make the process understandable by delving into the implementation specifics and outlining the fundamental concepts of music. Our effective model highlights the potential of GANs in music composition by producing cohesive music. After training for 50 epochs, the model exhibited a remarkable precision of 91.82 percent. This project uses the combination of Artificial Intelligence (AI) with music theory to provide intriguing new opportunities in the field of music. The model can be beneficial for different industries like gaming, music, entertainment, education, etc.

Acapella-based music generation with sequential models utilizing discrete cosine transform

Making musical instruments that accompany vocals in a song depends on the mood quality and the music composer’s creativity. The model created by other researchers has restrictions that include being limited to musical instrument digital interface files and relying on recurrent neural networks (RNN) or Transformers for the recursive generation of musical notes. This research offers the world’s first model capable of automatically generating musical instruments accompanying human vocal sounds. The model we created is divided into three types of sound input: short input, combed input, and frequency sound based on the discrete cosine transform (DCT). By combining the sequential models such as Autoencoder and gated recurrent unit (GRU) models, we will evaluate the performance of the resulting model in terms of loss and creativity. The best model has a performance evaluation that resulted in an average loss of 0.02993620155. The hearing test results from the sound output produced in the frequency range 0-1,600 Hertz can be heard clearly, and the tones are quite harmonious. The model has the potential to be further developed in future research in the field of sound processing.

Музыка в живописи Натальи Нестеровой

The article discusses the friendship and creative connections between the famous representatives of Russian culture — painter Natalya Nesterova and composer David Krivitsky. The relevance of the topic is connected with Nesterova’s 80th birthday anniversary the world is celebrating. This is a reason to draw attention to the fact that at the moment there is an extremely limited number of scientific works devoted to her heritage. This article for the first time publishes some unknown facts from her artist’s biography and presents her pictures from the composer’s personal archive. The research perspective is to reveal the significance of musical art and performance for N.I. Nesterova and to show the originality of her approach to depicting musicians and the intended symbolic meaning in her paintings with musical subjects. As an exceptional case, the article describes her activities as an illustrator for the music editions of the composer D.I. Krivitsky. It can be concluded that communication between the two bright creative personalities had a mutual influence and served as an impetus for the emergence of interesting works of art.

Age-related neural changes underlying long-term recognition of musical sequences

Aging is often associated with decline in brain processing power and neural predictive capabilities. To challenge this notion, we used magnetoencephalography (MEG) and magnetic resonance imaging (MRI) to record the whole-brain activity of 39 older adults (over 60 years old) and 37 young adults (aged 18–25 years) during recognition of previously memorised and varied musical sequences. Results reveal that when recognising memorised sequences, the brain of older compared to young adults reshapes its functional organisation. In fact, it shows increased early activity in sensory regions such as the left auditory cortex (100 ms and 250 ms after each note), and only moderate decreased activity (350 ms) in medial temporal lobe and prefrontal regions. When processing the varied sequences, older adults show a marked reduction of the fast-scale functionality (250 ms after each note) of higher-order brain regions including hippocampus, ventromedial prefrontal and inferior temporal cortices, while no differences are observed in the auditory cortex. Accordingly, young outperform older adults in the recognition of novel sequences, while no behavioural differences are observed with regards to memorised ones. Our findings show age-related neural changes in predictive and memory processes, integrating existing theories on compensatory neural mechanisms in non-pathological aging.

On the construal of MUSIC IS MOTION metaphor

Abstract TIME IS MOTION can be construed in two models. One is the Ego-based, the other is Time-based so as to form three time metaphors, MOVING TIME, TIME’S LANDSCAPE, and TEMPORAL SEQUENCE. When construing musical motion, people tend to borrow this conceptual framework and get three musical metaphors, i.e. MOVING MUSIC, MUSICAL LANDSCAPE, and MUSICAL SEQUENCE. The MUSICAL SEQUENCE metaphor falls into three categories, i.e. SEQUENCE IS THE RELATIVE POSITION ON THE TIME PATH, ON THE PITCH PATH, and ON THE KEY PATH. Time Metaphor and the Event Structure Metaphor jointly construct the two construal models for the metaphor MUSIC IS MOTION. This study tries to validate the previous research on conceptual construals in such a new field as music so as to enlarge the research range of the conceptual metaphor theory and provide reference for the subsequent research on music.

The Billboard Melodic Music Dataset (BiMMuDa)

We introduce the Billboard Melodic Music Dataset (BiMMuDa), which contains the lead vocal melodies of the top five songs of each year from 1950 to 2022 according to the Billboard year-end singles charts. In this article, the dataset’s compilation process and attributes are described in detail. The melody from each of the 371 songs was transcribed manually in full to create 371 MIDI (musical instrument digital interface) files, and then melodies from the songs’ different sections (e.g., verses, choruses) were exported into separate files to create an additional 1,133 MIDI files of shorter melodies. Lyrics to the songs are provided separately from the melodic transcriptions. This report includes comprehensive descriptions and graphical representations of the available metadata per song and per melody. Analysis of verse and chorus melodies revealed structural differences between them: chorus melodies have significantly fewer notes and lower note density, but larger melodic intervals on average. Whether added to existing datasets or used as a complete dataset, BiMMuDa can serve as ground truth data for a variety of MIR tasks as well as provide insight into the study of Western pop melody.

Spatiotemporal whole-brain activity and functional connectivity of melodies recognition.

Music is a non-verbal human language, built on logical, hierarchical structures, that offers excellent opportunities to explore how the brain processes complex spatiotemporal auditory sequences. Using the high temporal resolution of magnetoencephalography, we investigated the unfolding brain dynamics of 70 participants during the recognition of previously memorized musical sequences compared to novel sequences matched in terms of entropy and information content. Measures of both whole-brain activity and functional connectivity revealed a widespread brain network underlying the recognition of the memorized auditory sequences, which comprised primary auditory cortex, superior temporal gyrus, insula, frontal operculum, cingulate gyrus, orbitofrontal cortex, basal ganglia, thalamus, and hippocampus. Furthermore, while the auditory cortex responded mainly to the first tones of the sequences, the activity of higher-order brain areas such as the cingulate gyrus, frontal operculum, hippocampus, and orbitofrontal cortex largely increased over time during the recognition of the memorized versus novel musical sequences. In conclusion, using a wide range of analytical techniques spanning from decoding to functional connectivity and building on previous works, our study provided new insights into the spatiotemporal whole-brain mechanisms for conscious recognition of auditory sequences.

Sensorimotor Impairment in Aging and Neurocognitive Disorders: Beat Synchronization and Adaptation to Tempo Changes.

Understanding the nature and extent of sensorimotor decline in aging individuals and those with neurocognitive disorders (NCD), such as Alzheimer's disease, is essential for designing effective music-based interventions. Our understanding of rhythmic functions remains incomplete, particularly in how aging and NCD affect sensorimotor synchronization and adaptation to tempo changes. This study aimed to investigate how aging and NCD severity impact tapping to metronomes and music, with and without tempo changes. Patients from a memory clinic participated in a tapping task, synchronizing with metronomic and musical sequences, some of which contained sudden tempo changes. After exclusions, 51 patients were included in the final analysis. Participants' Mini-Mental State Examination scores were associated with tapping consistency. Additionally, age negatively influenced consistency when synchronizing with a musical beat, whereas consistency remained stable across age when tapping with a metronome. The results indicate that the initial decline of attention and working memory with age may impact perception and synchronization to a musical beat, whereas progressive NCD-related cognitive decline results in more widespread sensorimotor decline, affecting tapping irrespective of audio type. These findings underline the importance of customizing rhythm-based interventions to the needs of older adults and individuals with NCD, taking into consideration their cognitive as well as their rhythmic aptitudes.

Part of whose world? How The Little Mermaid (2023) attempts to revise the racist tropes of the 1989 animated film musical

The Little Mermaid (1989) has been the subject of much critical discussion but there has been little attention paid to its representation of race. Utilising textual analysis and drawing upon relevant critical paradigms from Disney studies and scholarship of film musicals, this article argues that The Little Mermaid was an implicitly racist narrative. Through analysis of the musical sequences this article shows how The Little Mermaid created a dichotomy between an exclusively white land life that is coded as superior to multiethnic Mer life. While white society on land is represented as postindustrial, well ordered and crime free, the multiethnic culture under the sea is coded not only as sexually unbridled, where promiscuous fish all “get the urge and start to play” with each other’s bodies, but as a thoroughly corrupt society racked with Welfare Queen coded villainesses dealing in black magic. The article contends that the 2023 remake attempts to revise many of the racist tropes of the original song sequences but, in its depiction of a society in which racial difference is devoid of socio-political significance, The Little Mermaid (2023)—especially in its Berkeley inspired dance sequences—reduces racial difference to a mere aesthetic of colour.

Composing Music for Choreographic Projects: Cases and Trends

All over the world the studies on the collaboration between music composers and dance creators are highly sought after. This is driven by the rapid evolution of musical and choreographic art fields in today’s creative industries. Organizing and promoting such cultural practices is crucial to complement the creative aspect. That’s why the present study examines the tendencies of compositional creativity in Kazakhstan in terms of its integration into choreographic projects. It can also be considered as a projection on cultural management. The article aims to the analytical review of the recent years’ implemented cases, that demonstrate the ways and formats of interaction between music and dance. The article considers the following questions: the interaction of music and choreography in a historical context; the creation of music and compilation of music for ballet; the design of music for dance and dance video, as well as the ways to support compositional creativity.This study of temporal and spatial art forms is built at the interface of several sciences. For the conceptual analysis the authors of the article required a wide range of historical and socio-cultural sources in the specified areas of art. The work was divided into several stages and used analytical, comparative and historical methods projected onto the Kazakh musical theater and other institutions related to the dance art. Also, an interviewing method was applied for the survey of practicing composers and directors, music editors and performers.The design of music and choreography has a wide range of relationships and demands that are necessary for the execution of audiovisual projects.This article with its understanding of cases in the field of ballet theater, dance troupes or video clips may be interesting both theoretically and practically not only for the scholars, but also for composers and choreographers who are only at the beginning of their creative paths. In addition, the article raises the problem of promoting and supporting compositional creativity. A thorough understanding of the music creation process for dance can foster more effective collaborations among creators, as well as the development of cultural management and clusters of creative industries.

MIDI-based generative neural networks with variational autoencoders for innovative music creation

By utilizing variational autoencoder (VAE) architectures in musical instrument digital interface (MIDI)-based generative neural networks (GNNs), this study explores the field of creative music composition. The study evaluates the success of VAEs in generating musical compositions that exhibit both structural integrity and a resemblance to authentic music. Despite achieving convergence in the latent space, the degree of convergence falls slightly short of initial expectations. This prompts an exploration of contributing factors, with a particular focus on the influence of training data variation. The study acknowledges the optimal performance of VAEs when exposed to diverse training data, emphasizing the importance of sufficient intermediate data between extreme ends. The intricacies of latent space dimensions also come under scrutiny, with challenges arising in creating a smaller latent space due to the complexities of representing data in N dimensions. The neural network tends to position data further apart, and incorporating additional information necessitates exponentially more data. Despite the suboptimal parameters employed in the creation and training process, the study concludes that they are sufficient to yield commendable results, showcasing the promising potential of MIDI-based GNNs with VAEs in pushing the boundaries of innovative music composition.

Comparative Symbolic Analysis of the Ethno-Fusion Genre: Insights and Perspectives

Abstract This study explores the integration of music and technology, illustrating their potential to collaboratively push the boundaries of musical exploration. Despite traditionally being viewed as unrelated, the combination of these two fields can significantly contribute to the progress of musical development. This study uses advanced computational methods to build a dataset filled with symbolic musical sequences that belong to a specific genre. This dataset is shown to be highly accurate and provides a detailed analysis of frequencies when examined closely, highlighting its quality and depth. We subject our dataset to comparative analysis with the renowned MAESTRO dataset, employing chromagrams to examine audio signals, rhythms, chords, solos, and note patterns in MIDI format through a variety of methods. This comparison underscores the superior quality of our sequences relative to those in the MAESTRO dataset, emphasizing the meticulousness of our sequence creation process. Moreover, we conduct internal evaluations of our dataset using both three-dimensional and two-dimensional approaches to melody representation, confirming its viability for future scholarly work. This effort seeks to enhance the music field by integrating computer science insights and methodologies, expanding the scope for future music technology research. It highlights the collaborative potential between musical creativity and technological advances in ongoing studies.