US Patent Issued to GOOGLE on Nov. 11 for "Targeted voice separation by speaker conditioned on spectrogram masking" (American, Swiss Inventors)

ALEXANDRIA, Va., Nov. 11 -- United States Patent no. 12,469,498, issued on Nov. 11, was assigned to GOOGLE LLC (Mountain View, Calif.).

"Targeted voice separation by speaker conditioned on spectrogram masking" was invented by Quan Wang (Hoboken, N.J.), Prashant Sridhar (New York), Ignacio Lopez Moreno (New York) and Hannah Muckenhim (Martigny, Switzerland).

According to the abstract* released by the U.S. Patent & Trademark Office: "Techniques are disclosed that enable processing of audio data to generate one or more refined versions of audio data, where each of the refined versions of audio data isolate one or more utterances of a single respective human speaker. Various implementations generate a refined version of audio data that isolat...