Synthetic Singers: A Review of Deep-Learning-based Singing Voice Synthesis Approaches

2026-01-20

AI Summary

This work presents the first comprehensive survey of deep learning–based singing voice synthesis, a field that has lacked systematic review. It organizes existing approaches into two major paradigms—cascaded and end-to-end—according to task formulation, and provides in-depth analysis of singing modeling and control techniques. The study also consolidates relevant datasets, annotation tools, and evaluation benchmarks. By structuring and synthesizing the core technical landscape, this survey not only fills a critical gap in the literature but also offers researchers and engineers a systematic reference. Furthermore, the authors release curated resources to support and accelerate community-wide progress in this emerging domain.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Recent advances in singing voice synthesis (SVS) have attracted substantial attention from both academia and industry. With the advent of large language models and novel generative paradigms, producing controllable, high-fidelity singing voices has become an attainable goal. Yet the field still lacks a comprehensive survey that systematically analyzes deep-learning-based singing voice synthesis systems and their enabling technologies. To address the aforementioned issue, this survey first categorizes existing systems by task type and then organizes current architectures into two major paradigms: cascaded and end-to-end approaches. Moreover, we provide an in-depth analysis of core technologies, covering singing modeling and control techniques. Finally, we review relevant datasets, annotation tools, and evaluation benchmarks that support training and assessment. In appendix, we introduce training strategies and further discussion of SVS. This survey provides an up-to-date review of the literature on SVS models, which would be a useful reference for both researchers and engineers. Related materials are available at https://github.com/David-Pigeon/SyntheticSingers.

Problem

• singing voice synthesis
• deep learning
• systematic survey
• generative models
• voice controllability

Innovation

• singing voice synthesis
• deep learning
• cascaded architecture
• end-to-end model
• voice control