US Patent Issued to Microsoft Technology Licensing on Feb. 3 for "Factorized neural transducer for multi-speaker speech recognition" (Washington Inventors)

ALEXANDRIA, Va., Feb. 3 -- United States Patent no. 12,542,135, issued on Feb. 3, was assigned to Microsoft Technology Licensing LLC (Redmond, Wash.).

"Factorized neural transducer for multi-speaker speech recognition" was invented by Jian Wu (Bellevue, Wash.), Jinyu Li (Bellevue, Wash.), Zhuo Chen (Woodinville, Wash.), Naoyuki Kanda (Bellevue, Wash.) and Takuya Yoshioka (Bellevue, Wash.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Systems and methods are provided for instantiating, modifying, adapting, and using a factorized neural transducer for multi-speaker automatic speech recognition. The factorized neural transducer includes a vocabulary predictor with multiple hidden states to process speech from d...