Intro
Preface
Organization
Contents
Music Technology in the IA Era
KANSEI Informatics and Music Technology in AI Era
1 Introduction
2 New Trend of Technology Development
3 KANSEI Informatics
4 AI and Music Technology
5 Conclusions
References
On Parallelism in Music and Language: A Perspective from Symbol Emergence Systems Based on Probabilistic Generative Models
1 Introduction
2 Language Acquisition and Music Composition Using PGMs
2.1 Multimodal Concept Formation and Lexical Acquisition in Robotics
2.2 Automatic Music Composition in Computers

3 Symbol Emergence Systems and Emergence of Semiotic Meanings
3.1 Symbol Emergence Systems
3.2 Collective Predictive Coding
4 On Parallelism in Music and Language
4.1 Parallelism on Syntax, Brain and Evolution
4.2 Symbol Emergence Systems on Music, Emotion, and Interoception
5 Conclusion
References
WaVAEtable Synthesis
1 Introduction
1.1 Motivation and Project Overview
1.2 Related Methods
2 WaVAEtable Synthesis
2.1 Sample Neural Network Architecture and Training
2.2 Wavetable Generation
2.3 Synthesizer Implementations

2.4 Incorporation of Existing Timbral Autoencoders
3 Future Work and Conclusion
References
Deep Learning-Based Music Instrument Recognition: Exploring Learned Feature Representations
1 Introduction
2 Attention-Based Model
2.1 Input Pre-processing
2.2 1st Stage: Post-processing and Representation
2.3 2nd Stage: CNN, Attention Mechanism and Classification
3 Datasets
4 Experimental Procedure
4.1 1st Stage: Pre-Training Objectives
4.2 2nd Stage: Downstream Instrument Recognition
5 Evaluation
6 Results and Discussion

6.1 Representation Post-processing: Impact on Performance
6.2 Learned Representations: Impact on Performance
7 Conclusions
References
Time-Span Tree Leveled by Duration of Time-Span
1 Introduction
2 Implementation Problems of Melodic-Morphing Algorithm
2.1 Ideas of Melodic Morphing
2.2 Partial Melody Reduction
2.3 Combining Two Melodies
2.4 Implementation Problems of Melodic-Morphing Algorithm
3 Implementation Problems of Deep-learning-based Time-Span Tree Analyzer
4 Solution: Time-Span Tree Leveled by Duration of Time Span
4.1 Automatic Melodic Morphing

4.2 Automatic Time-Span Tree Analysis by Deep Learning
5 Experiment and Results
5.1 Automating Melodic Morphing by Prioritization of Branches
5.2 Comparison of Seq2Seq and Transformer in Stepwise Time-Span Reduction
6 Conclusion
References
Evaluating AI as an Assisting Tool to Create Electronic Dance Music
1 Introduction
2 Related Work
3 Environment Setup and AI-Created Music
4 Test Structure and Efficiency Evaluation
4.1 Structure
4.2 Evaluation
5 Listening Evaluation
6 Summary and Conclusion
References
Interactive Systems for Music