US Patent Issued to Google on Feb. 18 for "Supervised and unsupervised training with contrastive loss over sequences" (New York, New Jersey, California Inventors)

ALEXANDRIA, Va., Feb. 19 -- United States Patent no. 12,230,249, issued on Feb. 18, was assigned to Google LLC (Mountain View, Calif.).

"Supervised and unsupervised training with contrastive loss over sequences" was invented by Andrew Rosenberg (Brooklyn, N.Y.), Bhuvana Ramabhadran (Mt. Kisco, N.Y.), Zhehuai Chen (Edgewater, N.J.), Yuan Wang (Brooklyn, N.Y.), Yu Zhang (Mountain View, Calif.) and Jesse Emond (Mountain View, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A method includes receiving audio data corresponding to an utterance and generating a pair of positive audio data examples. Here, each positive audio data example includes a respective augmented copy of the received audio data. For each...