High Quality Streaming Speech Synthesis with Low, Sentence-Length-Independent Latency.
Nikolaos EllinasGeorgios VamvoukakisKonstantinos MarkopoulosAimilios ChalamandarisGeorgia ManiatiPanos KakoulidisSpyros RaptisJune Sig SungHyoungmin ParkPirros TsiakoulisPublished in: CoRR (2021)
Keyphrases
- speech synthesis
- high quality
- speech recognition
- text to speech
- prosodic features
- vocal tract
- low quality
- real time
- data streams
- high resolution
- ground truth
- higher quality
- data mining
- streaming data
- response time
- natural language
- depth map
- language model
- high speed
- semantic role labeling
- pattern recognition
- bayesian networks