US 11,769,486 B2
System and method for data augmentation and speech processing in dynamic acoustic environments
Patrick A. Naylor, Reading (GB); Dushyant Sharma, Woburn, MA (US); Uwe Helmut Jost, Groton, MA (US); and William F. Ganong, III, Brookline, MA (US)
Assigned to Nuance Communications, Inc., Burlington, MA (US)
Filed by Nuance Communications, Inc., Burlington, MA (US)
Filed on Feb. 18, 2021, as Appl. No. 17/178,734.
Prior Publication US 2022/0262343 A1, Aug. 18, 2022
Int. Cl. G10L 15/06 (2013.01); G06N 20/00 (2019.01); G10L 25/03 (2013.01); H04R 1/22 (2006.01); H04R 1/40 (2006.01)
CPC G10L 15/063 (2013.01) [G06N 20/00 (2019.01); G10L 25/03 (2013.01); H04R 1/406 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A computer-implemented method, executed on a computing device, comprising:
defining a model representative of a plurality of acoustic variations to a speech signal, thus defining a plurality of time-varying spectral modifications, wherein defining the model representative of the plurality of acoustic variations to the speech signal includes defining a model representative of a plurality of acoustic variations to the speech signal dependent upon variations in a beampattern of an adaptive beamformer and movement of a plurality of sound sources relative to a beam of the adaptive beamformer; and
applying the plurality of time-varying spectral modifications to a plurality of feature coefficients of a target domain of a reference signal, thus generating a plurality of time-varying spectrally-augmented feature coefficients of the reference signal.