Review of end-to-end speech synthesis technology based on deep learning
arXiv:2104.09995
Abstract
As an indispensable part of modern human-computer interaction system, speech synthesis technology helps users get the output of intelligent machine more easily and intuitively, thus has attracted more and more attention. Due to the limitations of high complexity and low efficiency of traditional speech synthesis technology, the current research focus is the deep learning-based end-to-end speech synthesis technology, which has more powerful modeling ability and a simpler pipeline. It mainly consists of three modules: text front-end, acoustic model, and vocoder. This paper reviews the research status of these three parts, and classifies and compares various methods according to their emphasis. Moreover, this paper also summarizes the open-source speech corpus of English, Chinese and other languages that can be used for speech synthesis tasks, and introduces some commonly used subjective and objective speech quality evaluation method. Finally, some attractive future research directions are pointed out.
References in corpus (18)
- Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- NICE: Non-linear Independent Components Estimation
- Deep Voice: Real-time Neural Text-to-Speech
- Professor Forcing: A New Algorithm for Training Recurrent Networks
- Variational Lossy Autoencoder
- MelNet: A Generative Model for Audio in the Frequency Domain
- Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis
- Sliced Score Matching: A Scalable Approach to Density and Score Estimation
- CHiVE: Varying Prosody in Speech Synthesis with a Linguistically Driven Dynamic Hierarchical Conditional Variational Network
- JDI-T: Jointly trained Duration Informed Transformer for Text-To-Speech without Explicit Alignment
- Listening while Speaking: Speech Chain by Deep Learning
- Unified Mandarin TTS Front-end Based on Distilled BERT Model
- TalkNet: Fully-Convolutional Non-Autoregressive Speech Synthesis Model
- Deep Text-to-Speech System with Seq2Seq Model
- Effective parameter estimation methods for an ExcitNet model in generative text-to-speech systems
- Multi-Reference Neural TTS Stylization with Adversarial Cycle Consistency
- Exploiting Syntactic Features in a Parsed Tree to Improve End-to-End TTS