EMA2S: An End-to-End Multimodal Articulatory-to-Speech System.
Yu-Wen ChenKuo-Hsuan HungShang-Yi ChuangJonathan ShermanWen-Chin HuangXugang LuYu TsaoPublished in: CoRR (2021)
Keyphrases
- end to end
- multi stream
- speech recognition
- vocal tract
- audio visual
- speech signal
- speech synthesis
- hidden markov models
- automatic speech recognition
- multimodal interfaces
- formant frequencies
- wireless ad hoc networks
- ad hoc networks
- high bandwidth
- admission control
- internet protocol
- congestion control
- content delivery
- text to speech
- acoustic features
- application layer
- scalable video
- multipath