ALEXANDRIA, Va., Oct. 6 -- United States Patent no. 12,755,733, issued on Oct. 6, was assigned to Google LLC (Mountain View, Calif.).
"Tied and reduced RNN-T" was invented by Rami Botros (Mountain View, Calif.) and Tara Sainath (Jersey City, N.J.).
According to the abstract* released by the U.S. Patent & Trademark Office: "A RNN-T model includes a prediction network configured to, at each of a plurality of times steps subsequent to an initial time step, receive a sequence of non-blank symbols. For each non-blank symbol the prediction network is also configured to generate, using a shared embedding matrix, an embedding of the corresponding non-blank symbol, assign a respective position vector to the corresponding non-blank symbol, and weig...