Analyzing Input and Output Representations for Speech-Driven Gesture Generation.
Taras KucherenkoDai HasegawaGustav Eje HenterNaoshi KanekoHedvig KjellströmPublished in: CoRR (2019)
Keyphrases
- desired output
- input data
- multiple output
- multimodal interfaces
- speech recognition
- data driven
- text input
- input variables
- hidden markov models
- higher level
- finite state transducers
- neural network
- hand movements
- hidden units
- generation process
- speech signal
- data sets
- generation method
- feedback loop
- input device
- control signals
- gesture recognition
- bayesian networks
- dialogue system
- automatic speech recognition
- speech synthesis
- audio visual