Investigating Content-Aware Neural Text-to-Speech MOS Prediction Using Prosodic and Linguistic Features.
Alexandra VioniGeorgia ManiatiNikolaos EllinasJune Sig SungInchul HwangAimilios ChalamandarisPirros TsiakoulisPublished in: ICASSP (2023)
Keyphrases
- hyperspectral
- content aware
- text to speech
- linguistic features
- prosodic features
- text to speech synthesis
- speech synthesis
- structural features
- named entities
- spontaneous speech
- text classification
- semantic features
- named entity recognition
- linguistic knowledge
- sentence level
- feature set
- part of speech
- news stories
- pattern recognition
- machine learning
- content delivery
- text mining
- information extraction
- image processing