US Patent Issued to Google on Dec. 9 for "Speaker verification with multitask speech models" (California, New Jersey Inventors)

ALEXANDRIA, Va., Dec. 9 -- United States Patent no. 12,494,206, issued on Dec. 9, was assigned to Google LLC (Mountain View, Calif.).

"Speaker verification with multitask speech models" was invented by Alanna Foster Slocum (San Francisco), Yiling Huang (Edgewater, N.J.), Shelly Bensal (San Francisco) and Quan Wang (Hoboken, N.J.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A method includes obtaining a speaker identification (SID) model trained to predict speaker embeddings from utterances spoken by different speakers, the SID model includes a trained audio encoder and a trained SID head. The method also includes receiving a plurality of synthetic speech detection (SSD) training utterances that include a s...