Vincent Wan
21Patents
3h-index
26Co-inventors
59Inventor score
Filing activity: Mar 13, 2013 → Oct 17, 2023
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US9959368B2 | Computer generated emulation of a subject | Physics | 13 | Active |
| US9269347B2 | Text to speech system | Physics | 6 | Active |
| US11232780B1 | Predicting parametric vocoder parameters from prosodic features | Physics | 5 | Active |
| US9361722B2 | Synthetic audiovisual storyteller | Physics | 3 | Active |
| US9959657B2 | Computer generated head | Physics | 3 | Active |
| US11514888B2 | Two-level speech prosody transfer | Physics | 3 | Active |
| US11264010B2 | Clockwork hierarchical variational encoder | Physics | 2 | Active |
| US10923107B2 | Clockwork hierarchical variational encoder | Physics | 2 | Active |
| US12080272B2 | Attention-based clockwork hierarchical variational encoder | Physics | 1 | Active |
| US9454963B2 | Text to speech method and system using voice characteristic dependent weighting | Physics | 1 | Active |
| US11144597B2 | Computer generated emulation of a subject | Physics | 0 | Active |
| US12327544B2 | Two-level speech prosody transfer | Physics | 0 | Active |
| US11664011B2 | Clockwork hierarchal variational encoder | Physics | 0 | Active |
| US11790274B2 | Training neural networks to generate structured embeddings | Physics | 0 | Active |
| US11393453B2 | Clockwork hierarchical variational encoder | Physics | 0 | Active |
| US12260851B2 | Two-level text-to-speech systems using synthetic training data | Physics | 0 | Active |
| US11494695B2 | Training neural networks to generate structured embeddings | Physics | 0 | Active |
| US10373604B2 | Noise compensation in speaker-adaptive systems | Physics | 0 | Active |
| US11830474B2 | Predicting parametric vocoder parameters from prosodic features | Physics | 0 | Active |
| US12272349B2 | Attention-based clockwork hierarchical variational encoder | Physics | 0 | Active |
| US12125469B2 | Predicting parametric vocoder parameters from prosodic features | Physics | 0 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.