US Patent Issued to Google on March 24 for "End-to-end segmentation in a two-pass cascaded encoder automatic speech recognition model" (California, New Jersey Inventors)

ALEXANDRIA, Va., March 24 -- United States Patent no. 12,586,579, issued on March 24, was assigned to Google LLC (Mountain View, Calif.).

"End-to-end segmentation in a two-pass cascaded encoder automatic speech recognition model" was invented by Wenqian Ronny Huang (Mountain View, Calif.), Shuo-yiin Chang (Sunnyvale, Calif.), Tara N. Sainath (Jersey City, N.J.) and Yanzhang He (Mountain View, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A unified end-to-end segmenter and two-pass automatic speech recognition (ASR) model includes a first encoder, a first decoder, a second encoder, and a second decoder. The first encoder is configured to receive a sequence of acoustic frames and generate a first highe...