Sample-Efficient Representation Learning for Real-Time Latent Speaker State Characterization
A patent family: one line of work, grouped by real continuity rather than by a technology label
inferred from titles. 3 of these applications claim priority to application 63/061,018, filed Aug 4, 2020. A shared origin can cover several different claimed inventions, which is why members here may not share a title. In-force patents first, soonest term end first, then patent number. Expired patents follow, in the same order.
| Patent | Title | Holder | Status | Expires | Time remaining | Confidence |
|---|---|---|---|---|---|---|
| US 11,646,037 | Sample-Efficient Representation Learning for Real-Time Latent Speaker State Characterization In force · High confidence | Unity Technologies SF | In force | High confidence | ||
| US 11,790,921 | Speaker Separation Based on Real-Time Latent Speaker State Characterization In force · High confidence | Unity Technologies SF | In force | High confidence | ||
| US 12,315,516 | Speaker Separation Based on Real-Time Latent Speaker State Characterization In force · Medium confidence | Unity Technologies SF | In force | Medium confidence | ||
| Application 18/129,789 | Sample-Efficient Representation Learning for Real-Time Latent Speaker State Characterisation Abandoned Abandoned Mar 13, 2026 | Unity Technologies SF | Abandoned | Abandoned Mar 13, 2026 | not computed |
4 of 4 records shown. In-force patents first, soonest term end first, then patent number. Expired patents follow, in the same order.
3 of these are in force and held by Unity Technologies SF.
Updated Nov 15, 2026 · includes USPTO assignment & status data published Nov 15, 2026 and grant & fee data published Nov 10, 2026 (weekly) · how we know
Computed patent terms end on the dates shown. Related patents and pending applications may cover overlapping subject matter.