US Patent Issued to Google on March 18 for "One model unifying streaming and non-streaming speech recognition" (California, Washington Inventors)

ALEXANDRIA, Va., March 19 -- United States Patent no. 12,254,869, issued on March 18, was assigned to Google LLC (Mountain View, Calif.).

"One model unifying streaming and non-streaming speech recognition" was invented by Anshuman Tripathi (Mountain View, Calif.), Hasim Sak (Santa Clara, Calif.), Han Lu (Redmond, Wash.), Qian Zhang (Mountain View, Calif.) and Jaeyoung Kim (Cupertino, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A transformer-transducer model for unifying streaming and non-streaming speech recognition includes an audio encoder, a label encoder, and a joint network. The audio encoder receives a sequence of acoustic frames, and generates, at each of a plurality of time steps, a higher ...