Unified Speech and Gesture Synthesis Using Flow Matching.
Shivam MehtaRuibo TuSimon AlexandersonJonas BeskowÉva SzékelyGustav Eje HenterPublished in: ICASSP (2024)
Keyphrases
- matching algorithm
- multimodal interfaces
- hand movements
- speech recognition
- gesture recognition
- hidden markov models
- pattern matching
- matching process
- audio visual
- flow patterns
- spoken language
- unified model
- matching scheme
- texture synthesis
- speech signal
- graph matching
- keypoints
- multi modal
- pattern recognition
- sign language
- automatic speech recognition
- affine invariant
- shape matching
- flow field
- multi stream
- recognition engine
- endpoint detection