US Patent Issued to Google on Jan. 13 for "Unified end-to-end speech recognition and endpointing using a switch connection" (California, New Jersey Inventors)

ALEXANDRIA, Va., Jan. 13 -- United States Patent no. 12,525,228, issued on Jan. 13, was assigned to Google LLC (Mountain View, Calif.).

"Unified end-to-end speech recognition and endpointing using a switch connection" was invented by Shaan Jagdeep Patrick Bijwadia (San Francisco), Shuo-yiin Chang (Sunnyvale, Calif.), Bo Li (Fremont, Calif.), Yanzhang He (Palo Alto, Calif.), Tara N. Sainath (Jersey City, N.J.) and Chao Zhang (Mountain View, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A single E2E multitask model includes a speech recognition model and an endpointer model. The speech recognition model includes an audio encoder configured to encode a sequence of audio frames into corresponding higher-...