Ehsan Variani
15Patents
4h-index
34Co-inventors
56Inventor score
Filing activity: Mar 28, 2014 → Feb 27, 2024
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US9401148B2 | Speaker verification using neural networks | Physics | 42 | Active |
| US10140980B2 | Complex linear projection for acoustic modeling | Physics | 10 | Active |
| US10224058B2 | Enhanced multi-channel acoustic models | Physics | 9 | Active |
| US11062725B2 | Multichannel speech recognition using neural networks | Physics | 7 | Active |
| US11783849B2 | Enhanced multi-channel acoustic models | Physics | 1 | Active |
| US11568260B2 | Exponential modeling with deep learning features | Physics | 1 | Active |
| US12080283B2 | Multilingual re-scoring models for automatic speech recognition | Physics | 1 | Active |
| US11715458B2 | Efficient streaming non-recurrent on-device end-to-end model | Physics | 1 | Active |
| US10714078B2 | Linear transformation for speech recognition modeling | Physics | 0 | Active |
| US11393457B2 | Complex linear projection for acoustic modeling | Physics | 0 | Active |
| US12051404B2 | Efficient streaming non-recurrent on-device end-to-end model | Physics | 0 | Active |
| US12354598B2 | Rare word recognition with LM-aware MWER training | Physics | 0 | Active |
| US11922322B2 | Exponential modeling with deep learning features | Physics | 0 | Active |
| US12254875B2 | Multilingual re-scoring models for automatic speech recognition | Physics | 0 | Active |
| US12154581B2 | Cascaded encoders for simplified streaming and non-streaming ASR | Physics | 0 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.