StyleTTS 2: Towards Human-Level Text-to-Speech through Style Diffusion and Adversarial Training with Large Speech Language Models.
Yinghao Aaron LiCong HanVinay S. RaghavanGavin MischlerNima MesgaraniPublished in: CoRR (2023)
Keyphrases
- text to speech
- language model
- human level
- speech synthesis
- speech recognition
- language modeling
- artificial general intelligence
- machine intelligence
- intelligent systems
- document retrieval
- text to speech synthesis
- n gram
- prosodic features
- probabilistic model
- retrieval model
- human intelligence
- information retrieval
- statistical language models
- query expansion
- language modelling
- test collection
- web intelligence
- artificial intelligence
- cognitive science
- word error rate
- cognitive psychology
- document ranking
- relevance model
- automatic speech recognition
- language models for information retrieval
- word processing
- smoothing methods
- spoken term detection
- cognitive architecture
- speech signal
- information processing
- information retrieval systems
- decision making