Research

16 pages, 746 KiB

Open AccessArticle

Generating Fingerings for Piano Music with Model-Based Reinforcement Learning

by Wanxiang Gao, Sheng Zhang, Nanxi Zhang, Xiaowu Xiong, Zhaojun Shi and Ka Sun

Appl. Sci. 2023, 13(20), 11321; https://doi.org/10.3390/app132011321 - 15 Oct 2023

Viewed by 1474

The piano fingering annotation task refers to assigning finger labels to notes in piano sheet music. Good fingering helps improve the smoothness and musicality of piano performance. In this paper, we propose a method for automatically generating piano fingering using a model-based reinforcement [...] Read more.

The piano fingering annotation task refers to assigning finger labels to notes in piano sheet music. Good fingering helps improve the smoothness and musicality of piano performance. In this paper, we propose a method for automatically generating piano fingering using a model-based reinforcement learning algorithm. We treat fingering annotation as a partial constraint combinatorial optimization problem and establish an environment model for the piano performance process based on prior knowledge. We design a reward function based on the principle of minimal motion and use reinforcement learning algorithms to decide the optimal fingering combinations. Our innovation lies in establishing a more realistic environment model and adopting a model-based reinforcement learning approach, compared to model-free methods, to enhance the utilization of samples. We also propose a music score segmentation method to parallelize the fingering annotation task. The experimental section shows that our method achieves good results in eliminating physically impossible fingerings and reducing the amount of finger motion required in piano performance. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

14 pages, 283 KiB

Open AccessArticle

Unleashing Creative Synergies: A Mixed-Method Case Study in Music Education Classrooms

by Mário Anibal Cardoso, Elsa Maria Gabriel Morgado and Levi Leonido

Appl. Sci. 2023, 13(17), 9842; https://doi.org/10.3390/app13179842 - 31 Aug 2023

Cited by 1 | Viewed by 896

Abstract

Algorithmic music composition has been gaining prominence and recognition as an innovative approach to music education, providing students with opportunities to explore creativity, computational thinking, and musical knowledge. This study aims to investigate the impact of integrating algorithmic music composition in the classroom, [...] Read more.

Algorithmic music composition has been gaining prominence and recognition as an innovative approach to music education, providing students with opportunities to explore creativity, computational thinking, and musical knowledge. This study aims to investigate the impact of integrating algorithmic music composition in the classroom, examining its influence on student engagement, musical knowledge, and creative expression, as well as to enhance computational thinking skills. A mixed-method case study was conducted in three Basic Music Education classrooms in the north of Portugal, involving 71 participants (68 students and 3 music teachers). The results reveal: (i) both successes and challenges in integrating computational thinking concepts and practices; (ii) pedagogical benefits of integrating programming platforms, where programming concepts overlapped with music learning outcomes; and (iii) positive impact on participants’ programming self-confidence and recognition of programming’s importance. Integrating algorithmic music composition in the classroom positively influences student engagement, musical knowledge, and creative expression. The use of algorithmic techniques provides a novel and engaging platform for students to explore music composition, fostering their creativity, critical thinking, and collaboration skills. Educators can leverage algorithmic music composition as an effective pedagogical approach to enhance music education, allowing students to develop a deeper understanding of music theory and fostering their artistic expression. Future research should contribute to the successful integration of digital technologies in the Portuguese curriculum by further exploring the long-term effects and potential applications of algorithmic music composition in different educational contexts. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

28 pages, 1994 KiB

Open AccessArticle

TruMuzic: A Deep Learning and Data Provenance-Based Approach to Evaluating the Authenticity of Music

by Kuldeep Gurjar, Yang-Sae Moon and Tamer Abuhmed

Appl. Sci. 2023, 13(16), 9425; https://doi.org/10.3390/app13169425 - 19 Aug 2023

Viewed by 1227

Abstract

The digitalization of music has led to increased availability of music globally, and this spread has further raised the possibility of plagiarism. Numerous methods have been proposed to analyze the similarity between two pieces of music. However, these traditional methods are either focused [...] Read more.

The digitalization of music has led to increased availability of music globally, and this spread has further raised the possibility of plagiarism. Numerous methods have been proposed to analyze the similarity between two pieces of music. However, these traditional methods are either focused on good processing speed at the expense of accuracy or they are not able to properly identify the correct features and the related feature weights needed for achieving accurate comparison results. Therefore, to overcome these issues, we introduce a novel model for detecting plagiarism between two given pieces of music. The model does this with a focus on the accuracy of the similarity comparison. In this paper, we make the following three contributions. First, we propose the use of provenance data along with musical data to improve the accuracy of the model’s similarity comparison results. Second, we propose a deep learning-based method to classify the similarity level of a given pair of songs. Finally, using linear regression, we find the optimized weights of extracted features following the ground truth data provided by music experts. We used the main dataset, containing 3800 pieces of music, to evaluate the proposed method’s accuracy; we also developed several additional datasets with their own established ground truths. The experimental results show that our method, which we call ‘TruMuzic’, improves the overall accuracy of music similarity comparison by 10% compared to the other state-of-the-art methods from recent literature. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

18 pages, 1544 KiB

Open AccessArticle

Low Complexity Deep Learning Framework for Greek Orthodox Church Hymns Classification

by Lazaros Alexios Iliadis, Sotirios P. Sotiroudis, Nikolaos Tsakatanis, Achilles D. Boursianis, Konstantinos-Iraklis D. Kokkinidis, George K. Karagiannidis and Sotirios K. Goudos

Appl. Sci. 2023, 13(15), 8638; https://doi.org/10.3390/app13158638 - 27 Jul 2023

Viewed by 697

Abstract

The Byzantine religious tradition includes Greek Orthodox Church hymns, which significantly differ from other cultures’ religious music. Since the deep learning revolution, audio and music signal processing are often approached as computer vision problems. This work trains from scratch three different novel convolutional [...] Read more.

The Byzantine religious tradition includes Greek Orthodox Church hymns, which significantly differ from other cultures’ religious music. Since the deep learning revolution, audio and music signal processing are often approached as computer vision problems. This work trains from scratch three different novel convolutional neural networks on a hymns dataset to perform hymns classification for mobile applications. The audio data are first transformed into Mel-spectrograms and then fed as input to the model. To study in more detail our models’ performance, two state-of-the-art (SOTA) deep learning models were trained on the same dataset. Our approach outperforms the SOTA models both in terms of accuracy and their characteristics. Additional statistical analysis was conducted to validate the results obtained. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

21 pages, 9057 KiB

Open AccessArticle

Generative Music with Partitioned Quantum Cellular Automata

by Eduardo Reck Miranda and Hari Shaji

Appl. Sci. 2023, 13(4), 2401; https://doi.org/10.3390/app13042401 - 13 Feb 2023

Cited by 1 | Viewed by 2078

Abstract

Cellular automata (CA) are abstract computational models of dynamic systems that change some features with space and time. Music is the art of organising sounds in space and time, and it can be modelled as a dynamic system. Hence, CA are of interest [...] Read more.

Cellular automata (CA) are abstract computational models of dynamic systems that change some features with space and time. Music is the art of organising sounds in space and time, and it can be modelled as a dynamic system. Hence, CA are of interest to composers working with generative music. The art of generating music with CA hinges on the design of algorithms to evolve patterns of data and methods to render those patterns into musical forms. This paper introduces methods for creating original music using partitioned quantum cellular automata (PQCA). PQCA consist of an approach to implementing CA on quantum computers. Quantum computers leverage properties of quantum mechanics to perform computations differently from classical computers, with alleged advantages. The paper begins with some explanations of background concepts, including CA, quantum computing, and PQCA. Then, it details the PQCA systems that we have been developing to generate music and discusses practical examples. PQCA-generated materials for Qubism, a professional piece of music composed for London Sinfonietta, are included. The PQCA systems presented here were run on real quantum computers rather than simulations thereof. The rationale for doing so is also discussed. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

10 pages, 7323 KiB

Open AccessArticle

Q1Synth: A Quantum Computer Musical Instrument

by Eduardo Reck Miranda, Peter Thomas and Paulo Vitor Itaboraí

Appl. Sci. 2023, 13(4), 2386; https://doi.org/10.3390/app13042386 - 13 Feb 2023

Viewed by 2991

Abstract

This paper introduces Q1Synth, an unprecedented musical instrument that produces sounds from (a) quantum state vectors representing the properties of a qubit, and (b) its measurements. The instrument is presented on a computer screen (or mobile device, such as a tablet or smartphone) [...] Read more.

This paper introduces Q1Synth, an unprecedented musical instrument that produces sounds from (a) quantum state vectors representing the properties of a qubit, and (b) its measurements. The instrument is presented on a computer screen (or mobile device, such as a tablet or smartphone) as a Bloch sphere, which is a visual representation of a qubit. The performer plays the instrument by rotating this sphere using a mouse. Alternatively, a gesture controller can be used, e.g., a VR glove. While the sphere is rotated, a continuously changing sound is produced. The instrument has a ‘measure key’. When the performer activates this key, the instrument generates a program (also known as a quantum circuit) to create the current state vector. Then, it sends the program to a quantum computer over the cloud for processing, that is, measuring, in quantum computing terminology. The computer subsequently returns the measurement, which is also rendered into sound. Currently, Q1Synth uses three different techniques to make sounds: frequency modulation (FM), subtractive synthesis, and granular synthesis. The paper explains how Q1Synth works and details its implementation. A setup developed for a musical performance, Spinnings, with three networked Q1Synth instruments is also reported. Q1Synth and Spinnings are examples of how creative practices can open the doors to new application pathways for quantum computing technology. Additionally, they illustrate how such emerging technology is leading to new approaches to musical instrument design and musical creativity. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

18 pages, 4070 KiB

Open AccessArticle

An HMM-Based Approach for Cross-Harmonization of Jazz Standards

by Maximos Kaliakatsos-Papakostas, Konstantinos Velenis, Leandros Pasias, Chrisoula Alexandraki and Emilios Cambouropoulos

Appl. Sci. 2023, 13(3), 1338; https://doi.org/10.3390/app13031338 - 19 Jan 2023

Cited by 3 | Viewed by 1153

Abstract

This paper presents a methodology for generating cross-harmonizations of jazz standards, i.e., for harmonizing the melody of a jazz standard (Song A) with the harmonic context of another (Song B). Specifically, the melody of Song A, along with the chords that start and [...] Read more.

This paper presents a methodology for generating cross-harmonizations of jazz standards, i.e., for harmonizing the melody of a jazz standard (Song A) with the harmonic context of another (Song B). Specifically, the melody of Song A, along with the chords that start and end its sections (chord constraints), are used as a basis for generating new harmonizations with chords and chord transitions taken from Song B. This task involves potential incompatibilities between the components drawn from the two songs that take part in the cross-harmonization. In order to tackle such incompatibilities, two methods are introduced that are integrated in the Hidden Markov Model and the Viterbi algorithm. First, a rudimentary approach to chord grouping is presented that allows interchangeable utilization of chords belonging to the same group, depending on melody compatibility. Then, a “supporting” harmonic space of chords and probabilities is employed, which is learned from the entire dataset of the available jazz standards; this space provides local solutions when there are insurmountable conflicts between the melody and constraints of Song A and the harmonic context of Song B. Statistical and expert evaluation allow an analysis of the methodology, providing valuable insight about future steps. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

20 pages, 1440 KiB

Open AccessFeature PaperArticle

Induced Emotion-Based Music Recommendation through Reinforcement Learning

by Roberto De Prisco, Alfonso Guarino, Delfina Malandrino and Rocco Zaccagnino

Appl. Sci. 2022, 12(21), 11209; https://doi.org/10.3390/app122111209 - 04 Nov 2022

Cited by 4 | Viewed by 5862

Abstract

Music is widely used for mood and emotion regulation in our daily life. As a result, many research works on music information retrieval and affective human-computer interaction have been proposed to model the relationships between emotion and music. However, most of these works [...] Read more.

Music is widely used for mood and emotion regulation in our daily life. As a result, many research works on music information retrieval and affective human-computer interaction have been proposed to model the relationships between emotion and music. However, most of these works focus on applications in a context-sensitive recommendation that considers the listener’s emotional state, but few results have been obtained in studying systems for inducing future emotional states. This paper proposes Moodify, a novel music recommendation system based on reinforcement learning (RL) capable of inducing emotions in the user to support the interaction process in several usage scenarios (e.g., games, movies, smart spaces). Given a target emotional state, and starting from the assumption that an emotional state is entirely determined by a sequence of recently played music tracks, the proposed RL method is designed to learn how to select the list of music pieces that better “match” the target emotional state. Differently from previous works in the literature, the system is conceived to induce an emotional state starting from a current emotion instead of capturing the current emotion and suggesting certain songs that are thought to be suitable for that mood. We have deployed Moodify as a prototype web application, named MoodifyWeb. Finally, we enrolled 40 people to experiment MoodifyWeb, employing one million music playlists from the Spotify platform. This preliminary evaluation study aimed to analyze MoodifyWeb’s effectiveness and overall user satisfaction. The results showed a highly rated user satisfaction, system responsiveness, and appropriateness of the recommendation (up to 4.30, 4.45, and 4.75 on a 5-point Likert, respectively) and that such recommendations were better than they thought before using MoodifyWeb (6.45 on a 7-point Likert). Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

13 pages, 1567 KiB

Open AccessArticle

Annotated-VocalSet: A Singing Voice Dataset

by Behnam Faghih and Joseph Timoney

Appl. Sci. 2022, 12(18), 9257; https://doi.org/10.3390/app12189257 - 15 Sep 2022

Cited by 1 | Viewed by 3189

Abstract

There are insufficient datasets of singing files that are adequately annotated. One of the available datasets that includes a variety of vocal techniques (n = 17) and several singers (m = 20) with several WAV files (p = 3560) is [...] Read more.

There are insufficient datasets of singing files that are adequately annotated. One of the available datasets that includes a variety of vocal techniques (n = 17) and several singers (m = 20) with several WAV files (p = 3560) is the VocalSet dataset. However, although several categories, including techniques, singers, tempo, and loudness, are in the dataset, they are not annotated. Therefore, this study aims to annotate VocalSet to make it a more powerful dataset for researchers. The annotations generated for the VocalSet audio files include fundamental frequency contour, note onset, note offset, the transition between notes, note F0, note duration, Midi pitch, and lyrics. This paper describes the generated dataset and explains our approaches to creating and testing the annotations. Moreover, four different methods to define the onset/offset are compared. Full article

(This article belongs to the Special Issue Algorithmic Music and Sound Computing)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Algorithmic Music and Sound Computing

Share This Special Issue

Special Issue Editor

Special Issue Information

Published Papers (9 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI