Jia Cui
10Patents
1h-index
7Co-inventors
39Inventor score
Filing activity: Aug 30, 2018 → Nov 17, 2022
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US11004443B2 | Multistage curriculum training framework for acoustic-to-word speech recognition | Physics | 4 | Active |
| US11037547B2 | Token-wise training for attention based end-to-end speech recognition | Physics | 1 | Active |
| US11972754B2 | Multi-task training architecture and strategy for attention-based speech recognition system | Physics | 0 | Active |
| US11257481B2 | Multi-task training architecture and strategy for attention-based speech recognition system | Physics | 0 | Active |
| US10861441B2 | Large margin training for attention-based end-to-end speech recognition | Physics | 0 | Active |
| US11803618B2 | N-best softmax smoothing for minimum bayes risk training of attention based sequence-to-sequence models | Physics | 0 | Active |
| US11636848B2 | Token-wise training for attention based end-to-end speech recognition | Physics | 0 | Active |
| US11551136B2 | N-best softmax smoothing for minimum bayes risk training of attention based sequence-to-sequence models | Physics | 0 | Active |
| US10923117B2 | Best path change rate for unsupervised language model weight selection | Physics | 0 | Active |
| US10672382B2 | Input-feeding architecture for attention based end-to-end speech recognition | Physics | 0 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.