Artificial Intelligence in Music, Sound, Art and Design

Name: Artificial Intelligence in Music, Sound, Art and Design | 11th International Conference, EvoMUSART 2022, Held as Part of EvoStar 2022, Madrid, Spain, April 20-22, 2022, Proceedings
Brand: Springer
Price: 128.39 EUR
Availability: OnlineOnly

11th International Conference, EvoMUSART 2022, Held as Part of EvoStar 2022, Madrid, Spain, April 20-22, 2022, Proceedings

Tiago Martins Nereida Rodríguez-Fernández Sérgio M. Rebelo(Herausgeber*in)

Springer (Verlag)

Erschienen am 15. April 2022

XI, 417 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-031-03789-4 (ISBN)

128,39 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents
Long Talks
SonOpt: Sonifying Bi-objective Population-Based Optimization Algorithms
1 Introduction
2 Related Work
3 Methodology and System Overview
4 Experimental Study
4.1 Experimental Setup
4.2 Experimental Analysis
5 Conclusion and Future Work
References
A Systematic Evaluation of GPT-2-Based Music Generation
1 Introduction
2 Background
2.1 GPT-2 Models
2.2 Representing Music Data for GPT-2 Input
2.3 Statistical Analysis of Generative Music Models
3 Musical Metrics
4 Dataset Curation
5 Evaluating Generative Model Output
5.1 Varying the Training Level
5.2 Varying the Training Corpus
6 Web Application
7 Conclusions and Future Work
References
Expressive Aliens - Laban Effort Factors for Non-anthropomorphic Morphologies
1 Introduction
2 Background
2.1 Movement Qualities
2.2 Motion Synthesis
2.3 Anthropomorphic versus Non-anthropomorphic Characters
2.4 Physical Validity versus Expressivity
3 Implementation
3.1 Morphologies
3.2 SAC Algorithm
3.3 Observation Vector
3.4 Rewards
4 Training
5 Results
6 Discussion
7 Conclusion and Outlook
References
Painting with Evolutionary Algorithms
1 Introduction
2 Rearranging Brush Strokes
3 Algorithms
4 Experiment and Results
5 Extrapolation
6 Some Final Remarks
References
Evolutionary Construction of Stories that Combine Several Plot Lines
1 Introduction
2 Related Work
2.1 Plot Line Combination
2.2 Computational Metrics for Stories
2.3 Evolutionary Construction of Narratives
3 An Evolutionary Multiplot Story Composer
3.1 The Knowledge Resources
3.2 Character Fusion and Discourse Planning
3.3 Representing Multiplot Stories for Evolutionary Construction
3.4 Constructing an Initial Population
3.5 Evolutionary Operators
3.6 Fitness Functions
4 Discussion
4.1 Results
4.2 Relation with Previous Work
5 Conclusions
References
Fashion Style Generation: Evolutionary Search with Gaussian Mixture Models in the Latent Space
1 Introduction
2 Related Work
2.1 GANs
2.2 Fashion Styles
2.3 Evolutionary Search of GANs' Latent Space
3 Dataset
4 Model
4.1 Generative Model
4.2 Style Model
4.3 Evolutionary Search
5 Results
6 Discussion
7 Conclusion and Future Work
References
Classification of Guitar Effects and Extraction of Their Parameter Settings from Instrument Mixes Using Convolutional Neural Networks
1 Introduction
2 Method and Materials
2.1 Dataset for Guitar Effect Parameter Extraction
2.2 Dataset for Guitar Effect Classification
2.3 Time-Frequency Representations
2.4 Convolutional Neural Networks
2.5 Training and Evaluation
2.6 Baseline
2.7 Robustness Analysis
3 Results
3.1 Effect Classification
3.2 Effect Parameter Extraction
3.3 Robustness to Noise and Pitch Shifts
4 Discussion
4.1 Guitar Effect Classification
4.2 Guitar Effect Parameter Extraction
4.3 Robustness
4.4 Limitations
5 Conclusion
References
Aesthetic Evaluation of Experimental Stimuli Using Spatial Complexity and Kolmogorov Complexity
1 Introduction
2 Conceptual Model
3 Spatial Complexity Measure
4 Kolmogorov Complexity of 2D Patterns
5 Experiment and Results
5.1 Method
5.2 Material
5.3 Procedure
5.4 Results
5.5 Procedure for the Extended Study
5.6 Results and Analysis
6 Discussions
References
Towards the Generation of Musical Explanations with GPT-3
1 Introduction
2 Background
2.1 Communication in Human-Machine Music Interactions
2.2 Transformer-Based Approaches in Music
2.3 Transformer-Based Approaches in Explainable AI
3 Musical Capability of GPT-3
3.1 Extracting the Key from a Sequence of Notes
3.2 Providing Explanations for a Fictional Song
3.3 Extracting Musical Information Using MusicABC Notation
4 Explaining Musical Decisions
4.1 Methodology
4.2 Results
5 Discussion
6 Conclusion
References
Lamuse: Leveraging Artificial Intelligence for Sparking Inspiration
1 Introduction
2 The Creative Process
2.1 Context
2.2 General Assumptions and Process
2.3 Data Requirements
3 The Artificial Intelligence
3.1 Artwork Decomposition
3.2 Recomposing with a Visual Universe and Projection
3.3 Style Transfer
4 Results Analysis and Discussion
4.1 E. Potier
4.2 Rarès-Victor
4.3 N. Varoqui
4.4 O. Masmonteil
5 Conclusion
References
EvoDesigner: Towards Aiding Creativity in Graphic Design
1 Introduction
2 Related Work
3 Approach
3.1 Evolutionary Engine
4 Experimental Setup and Results
5 Conclusion
References
Conditional Drums Generation Using Compound Word Representations
1 Introduction
2 Related Work
3 Data Encoding Representation
3.1 Encoder Representation - Conditional Information
3.2 Decoder Representation - Generated Drum Sequences
4 Proposed Architecture
4.1 Encoder - Decoder
4.2 Implementation Details
5 Experimental Setup
5.1 Dataset and Preprocessing
5.2 Evaluation Metrics
6 Results
6.1 Objective Evaluation
6.2 Subjective Evaluation
7 Conclusions
References
Music Style Transfer Using Constant-Q Transform Spectrograms
1 Introduction
2 Related Work
3 Experimental Setup
4 Experiments and Results
4.1 CycleGAN Hyperparameter Experiment
4.2 CQTGAN Sample Rates Experiment
4.3 Unseen Audio Examples Experiment
5 Survey Evaluation
6 Discussion
7 Conclusions and Future Work
References
SpeechTyper: From Speech to Typographic Composition
1 Introduction
2 Related Work
3 SpeechTyper
3.1 Extracting Speech Data
3.2 Designing Glyphs
3.3 Creating Typographic Compositions Based on Speech
4 Experimentation
4.1 Setup
4.2 Results/Discussion
5 Conclusion and Future Work
References
A Creative Tool for the Musician Combining LSTM and Markov Chains in Max/MSP
1 Introduction
2 Data Representation
2.1 Pitch
2.2 Time
2.3 MIDI Analyzer
3 LSTM Data Encoding
3.1 Pitch
3.2 Rhythm
3.3 Machine Learning Datasets
4 Scramble
5 User Interface
6 Experimental Results
7 Conclusions and Future Work
References
Translating Emotions from EEG to Visual Arts
1 Introduction
2 Background and Related Works
3 Preparation of Datasets
4 Pipeline
4.1 Extra Losses
5 Experiments and Results
5.1 Example Experiment
6 Results Assessment: Online Survey
7 Discussion
8 Conclusions
References
Emotion-Driven Interactive Storytelling: Let Me Tell You How to Feel
1 Introduction
2 Background
3 Methodology
3.1 Emotion Recognition
3.2 Interface Design
3.3 System's Assessment
4 Results
5 Discussion
6 Conclusion
References
Modern Evolution Strategies for Creativity: Fitting Concrete Images and Abstract Concepts
1 Introduction
2 Background
3 Modern Evolution Strategies for Creativity
4 Fitting Concrete Target Image
5 Fitting Abstract Concept with CLIP
6 Discussion and Conclusion
References
Co-creative Product Design with Interactive Evolutionary Algorithms: A Practice-Based Reflection
1 Introduction
2 Background
3 Related Work
4 Design Study
4.1 Design Task
4.2 Study Method
4.3 Participant: The Designer
4.4 Software Tools and Algorithm
4.5 Findings: Introspective Design Reflections
5 Discussion
5.1 Early Constraining of the Design Space
5.2 Support in Problem-Solution Co-evolution
5.3 Escaping and Falling in a Fixation Trap
5.4 The IGA as Creative Partner
5.5 Study Limitations
6 Conclusion and Future Work
References
Sound Model Factory: An Integrated System Architecture for Generative Audio Modelling
1 Background and Motivation
1.1 Previous Work
2 Architecture
2.1 System Overview
2.2 System Components
3 Connecting the GAN and the RNN
3.1 Parameter Linearization
4 Evaluation
4.1 Human Evaluation Adaptively Smoothed Latent Space
4.2 Parameter Response Time
4.3 Sound Quality Evaluation Based on Audio Classification
4.4 Continuous Interpolation of Pitch and Timbre
5 Conclusion
References
Short Talks
An Application of Neural Embedding Models for Representing Artistic Periods
1 Introduction
2 Data
3 Methodology
3.1 word2vec
3.2 t-SNE
4 Results
4.1 Salvador Dalí
4.2 Vincent van Gogh
4.3 Pablo Picasso
5 Discussion
6 Conclusion
References
MusIAC: An Extensible Generative Framework for Music Infilling Applications with Multi-level Control
1 Introduction
2 Related Work
3 Proposed Model and Representation
3.1 Adding Control Features
3.2 Data Representation
3.3 Model Architecture
4 Experimental Setup
4.1 Dataset
4.2 Model Configuration and Training
4.3 Inference Strategy
5 Evaluation
5.1 Objective Evaluation Using Selected Metrics
5.2 The Interactive Interface and Controllability
6 Conclusion
References
A Study on Noise, Complexity, and Audio Aesthetics
1 Introduction
2 Conceptual over Perceptual
2.1 Japanoise
3 Complexity
3.1 Complexity and Computational Aesthetics
4 Proposed Aesthetic Metrics
5 Experiment
6 Discussion
References
Quality-Diversity for Aesthetic Evolution
1 Introduction
2 Quality-Diversity Search
3 Generative System
4 Evolutionary Method
4.1 Fitness
4.2 Diversity
4.3 MAP-Elites Implementation
5 Experiments and Results
6 Discussion
7 Conclusion
References
Classifying Biometric Data for Musical Interaction Within Virtual Reality
1 Introduction
2 Literature Review
3 Research Methodology
3.1 Technical Hardware
3.2 Technical Software and Processes
3.3 Data Processing and Machine Learning
3.4 Performance Gestures Used
4 Results
4.1 Continuous Model Accuracies
4.2 Classifier Model Accuracies
4.3 Application to Interactive Music Composition Within VR
5 Conclusions
References
Generating Novel Furniture with Machine Learning
1 Introduction
2 Background
2.1 A Graph-Based Design Language
2.2 Deep Neural Nets (DNN)
3 The Architecture of a Generative Design Tool
3.1 Representing (Product) Structure as a Graph
3.2 The Use of Deep Neural Nets
4 The Generation of Novel Graph Structures
4.1 Object Analysis and Conversion to Graph
4.2 Conversion of the Graphs to Training Data
4.3 Training the DNN (Transformer) Model
4.4 Using the Transformer's Latent Space Variables to Train a VAE
4.5 The Generation of Novel Graphs Using the transformer's Decoder
4.6 Converting the Output String Representation to a Graph and Image
4.7 Building the Machine Learning Architecture
5 Outcomes and Evaluation
5.1 Generation of Novel Graphs
5.2 Measuring the Extent of Novelty
5.3 Selecting and Refining Generated Designs
6 Conclusion
References
Correction to: SpeechTyper: From Speech to Typographic Composition
Correction to: Chapter "SpeechTyper: From Speech to Typographic Composition" in: T. Martins et al. (Eds.): Artificial Intelligence in Music, Sound, Art and Design, LNCS 13221, https://doi.org/10.1007/978-3-031-03789-4_14
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken