US Patent Issued to Amazon Technologies on March 18 for "Augmenting datasets for training audio generation models" (British, Polish, German, Spanish Inventors)

ALEXANDRIA, Va., March 19 -- United States Patent no. 12,254,864, issued on March 18, was assigned to Amazon Technologies Inc. (Seattle).

"Augmenting datasets for training audio generation models" was invented by Mateusz Aleksander Lajszczak (Cambridge, Great Britain), Adam Marek Gabrys (Sopot, Poland), Arent van Korlaar (London), Ruizhe Li (London), Elena Sergeevna Sokolova (London), Jaime Lorenzo Trueba (Madrid), Arnaud Vincent Pierre Yves Joly (Cambridge, Great Britain), Marco Nicolis (London) and Ekaterina Petrova (Oberhaching, Germany).

According to the abstract* released by the U.S. Patent & Trademark Office: "A target voice dataset may be augmented using speech prediction. Encoder and decoder models may be trained to encode audio d...