EMA2S: An End-to-End Multimodal Articulatory-to-Speech System.
Yu-Wen ChenKuo-Hsuan HungShang-Yi ChuangJonathan ShermanWen-Chin HuangXugang LuYu TsaoPublished in: ISCAS (2021)
Keyphrases
- end to end
- multi stream
- speech recognition
- vocal tract
- audio visual
- speech signal
- speech synthesis
- multimodal interfaces
- hidden markov models
- smart card
- multipath
- ad hoc networks
- automatic speech recognition
- high bandwidth
- congestion control
- wireless ad hoc networks
- admission control
- formant frequencies
- text to speech
- acoustic features
- rate allocation
- web services
- content delivery
- language model
- visual speech