US Patent Issued to Google on March 25 for "Two-level text-to-speech systems using synthetic training data" (British, American, Japanese Inventors)

ALEXANDRIA, Va., March 26 -- United States Patent no. 12,260,851, issued on March 25, was assigned to Google LLC (Mountain View, Calif.).

"Two-level text-to-speech systems using synthetic training data" was invented by Lev Finkelstein (Mountain View, Calif.), Chun-an Chan (Mountain View, Calif.), Byungha Chun (Tokyo), Norman Casagrande (London), Yu Zhang (Mountain View, Calif.), Robert Andrew James Clark (Hertfordshire, Great Britain) and Vincent Wan (London).

According to the abstract* released by the U.S. Patent & Trademark Office: "A method includes obtaining training data including a plurality of training audio signals and corresponding transcripts. Each training audio signal is spoken by a target speaker in a first accent/dialect. Fo...