Emotional End-to-End Neural Speech Synthesizer

2017/11/15 by Lee, Younggun, Rabiee, Azam, Lee, Soo-Young · 4 citations
#Audio and Speech Processing (eess.AS) #Computation and Language (cs.CL) #FOS: Computer and information sciences #FOS: Electrical engineering #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.1711.05447

Abstract

In this paper, we introduce an emotional speech synthesizer based on the recent end-to-end neural model, named Tacotron. Despite its benefits, we found that the original Tacotron suffers from the exposure bias problem and irregularity of the attention alignment. Later, we address the problem by utilization of context vector and residual connection at recurrent neural networks (RNNs). Our experiments showed that the model could successfully train and generate speech for given emotion labels.

Cited by

• EMORL-TTS: Reinforcement Learning for Fine-Grained Emotion Control in LLM-based TTS
• LibriTTS-VI: A Public Corpus and Novel Methods for Efficient Voice Impression Control
• TTS-CtrlNet: Time varying emotion aligned text-to-speech generation with ControlNet
• Prompt-Unseen-Emotion: Zero-shot Expressive Speech Synthesis with Prompt-LLM Contextual Knowledge for Mixed Emotions

Related

• Adjusting Pleasure-Arousal-Dominance for Continuous Emotional\n Text-to-speech Synthesizer
• Generative Emotional AI for Speech Emotion Recognition: The Case for Synthetic Emotional Speech Augmentation
• Controllable Emotion Transfer For End-to-End Speech Synthesis
• GANtron: Emotional Speech Synthesis with Generative Adversarial Networks
• End-to-End Emotional Speech Synthesis Using Style Tokens and\n Semi-Supervised Training